A Thesis Submitted for the Degree of PhD at the University of St. Andrews


Full metadata for this item is available in the St Andrews Digital Research Repository at:

Please use this identifier to cite or link to this item:

This item is protected by original copyright

Intra-sexual competition and vocal counterstrategies in wild female chimpanzees (Pan troglodytes schweinfurthii)

Simon W. Townsend

Dissertation submitted to the University of St Andrews for the degree of Doctor of Philosophy August 2008

I, Simon W. Townsend, hereby certify that this thesis, which is approximately 40,000 words in length, has been written by me, that it is the record of work carried out by me and that it has not been submitted in any previous application for a higher degree. I was admitted as a research student in October, 2005 and as a candidate for the degree of Doctor of Philosophy in October, 2006; the higher study for which this is a record was carried out in the University of St Andrews between 2005 and 2008. Date……… Signature of candidate………………………

I hereby certify that the candidate has fulfilled the conditions of the Resolution and Regulations appropriate for the degree of Doctor of Philosophy in the University of St Andrews and that the candidate is qualified to submit this thesis in application for that degree. Date………. Signature of supervisor……………………….. In submitting this thesis to the University of St Andrews I understand that I am giving permission for it to be made available for use in accordance with the regulations of the University Library for the time being in force, subject to any copyright vested in the work not being affected thereby. I also understand that the title and the abstract will be published, and that a copy of the work may be made and supplied to any bona fide library or research worker, that my thesis will be electronically accessible for personal or research use unless exempt by award of an embargo and that the library has the right to migrate my thesis into new electronic forms as required to ensure continued access to the thesis. I have obtained any third-party copyright permissions that may be required in order to allow such access and migration. Date………… Signature of candididate……………………

Marietta. it just happened that way. whilst I did have to sit and wait for you to finish “horsing” too many times. My gratitude to the other St Andrews people: Jenny. guidance. support and fondues will always be remembered. I am extremely grateful to my parents – you trusted my judgement and were always around at the most crucial times. from the forest to the wolfson. Ugandan Wildlife Authority. it would not have been possible without the invaluable contribution of my supervisor Professor(!) Klaus Zuberbuhler. My thanks to you in so many ways. I am grateful to the Biotechnology and Biological Science Research Council for fully funding this PhD and to the Presidents Office. you patiently listened to my questions and always answered with such clarity. Dave Perrett. Most importantly. Anne. Katie. Fee: thanks for pretending to fall asleep whenever we talked research. thanks for making me get on that Kenya Airways flight. Andy Whiten. His encouragement. your positivity and mockery made the three years a lot easier. Uganda. Bro. Thanks also to Tobias Deschner. I hope the Wolfson sex-ratio won’t always be so female-skewed. Jamie. Thanks to Zarin Machanda. Ugandan National Council for Science and Technology and Fred Babweteera for permission to live and work in the Budongo Forest. April. Zanna. Peter Slater. Nick Newton-Fisher and Dora Biro for nurturing my interest in chimpanzees and for helping me through those first 8 weeks and Vernon Reynolds for giving me the opportunity to study such a fascinating animal in the first place. . Marion and Emilie. First and foremost. I’m sure it stopped me from getting too serious. Kate Arnold and Melissa Emery Thompson for always being available and for their everhelpful comments. Vincent Janik. I didn’t mean to follow you.Acknowledgements I’d like to take this opportunity to thank a number of people for their help towards this thesis. Thanks to Dick Byrne for his continual constructive criticisms and insightful discussions and for reminding me that 6 million years really isn’t that much. Lucy.

It was a privilege to work with you. . my sincerest gratitude to Monday M. most notably Sam Adiue.Last but by no means least. Gophine (Tuesday) Geresomu (his names sake) and Stephen Amati. Jackson (Mvala). Gideon and the other field assistants who helped me over the 16 months in Uganda. this thesis is as much yours as it is mine.

39 Infanticide 3 ……………………………………………………………….. 5 Female competition in chimpanzees ……………………………………………...72 Statistics……………………………………………………………………74 Results Female identity 75 ……………………………………………………………… Hormone analyses………………………………………………………….75 Male dominance rank………………………………………………………...54 Results Mating behaviour of chimpanzees…………………………………………57 Hormone analyses………………………………………………………….14 Copulation calls ……………………………………………………………17 Victim Screams……………………………………………………………22 Outlook…………………………………………………………………… 25 Chapter two: Methods Study site and subjects……………………………………………………… 27 Data collection………………………………………………………………31 Chapter three: Lethal female competition in wild chimpanzees Introduction……………………………………………………………… 35 Methods…………………………………………………………………… 37 Results Infanticide 1………………………………………………………………. 71 Acoustic analyses………………………………………………………….53 Behavioural observations………………………………………………….60 Audience effects……………………………………………………………60 Discussion …………………………………………………………………. 37 Infanticide 2………………………………………………………………... hormone analyses. behavioural observations………………….76 Discussion…………………………………………………………………. 79 ..7 The role of vocalisations………………………………………………….63 Chapter five: Copulation calls in female chimpanzees convey social information Introduction………………………………………………………………… 68 Methods Study site. 40 …………………………………………………………………. 51 Hormone analysis…………………………………………………………...... Methods Copulation calls…………………………………………………………...42 Discussion Chapter four: Female chimpanzees use copulation calls flexibly to prevent social competition Introducion……………………………………………………………… 48 ……..Chapter one: General Introduction Background Origins of female competition…………………………………………… 3 Chimpanzee socioecology ……………………………………………….

..…..……….95 Results Playback responses ………………………………………………………….Chapter six: Wild chimpanzees distinguish between different scream types: evidence from a playback study Introduction……………………………………………………………… 85 Methods Playback stimuli………………………………………………………..91 Playback trials and data analysis………………………………….89 Protocol……………………………………………………………………. References…………………………………………………………………. ………………………………99 Short-term behavioural changes to playbacks Long-term behavioural changes to playbacks ………………………………100 Discussion 101 …………………………………………………………………… Chapter seven: General discussion Female competition in Budongo chimpanzees……………………. 148 ..126 Conclusion …………………………………………………………………128 .130 Appendix I………………………………………………………………….96 Effects of prior scream exposure 99 ……………………………………………..……… 106 Chimpanzee calls as social tools …………………………………………… 109 Other animal examples……………………………………………………… 120 Chimpanzee communication and cognition………………………………… 123 Future directions…………………………………………………………….

Table 6.1 Results of Univariate ANOVA’s comparing each acoustic parameter measured during POP and Post-POP. in each condition.1 The adult and sub-adult males of the Sonso community present during the 14 month study period. Mean values across the 6 subjects and the standard deviations are shown. Table 6. Uganda.2 The functional hypotheses so far presented to explain the adaptive function of primate copulation calls.3 Behavioural responses of subjects to each of the three playback conditions. Table 5.1 The mean values and SD of 4 acoustic measures for each scream condition.2 The mean duration and RMS amplitude of stimuli in each condition.1 Adult males. Table 2.1 The 14 different chimpanzee call variants and their contexts of production. Table 4.2 The adult females. 149 . Table 6. 16 18 29 30 56 58 76 90 91 97 99 Appendix table 1 The mean values and SD of four acoustic measures for mild and severe scream conditions. Table 4. before playback.2 Rank and copulation calling behaviour of seven adult females of the Sonso community. Table 1. their dependent offspring and adolescent females of the Sonso community during the study period. Table 2.List of tables Table 1.4 The mean and SD of screaming bout duration witnessed and heard. Budongo Forest. Table 6. dominance value scores and rank groups for the adult males of the Sonso community over the study duration.

4 Time-frequency spectrogram for female chimpanzees (i) LL and (ii) WL when copulating with high and low-ranked adult males. over the last decade. centred at the core of the Sonso community’s home range.3 Map showing the main home range of the Sonso community and the grid system covering it. 27 28 31 42 52 54 58 59 61 Figure 4.6 Line graphs showing the mean number of high and equal ranked females in the audience when copulating with high ranked males and low ranked males. Figure 4. Figure 2. Figure 6. Figure 5.2 Map of the Budongo forest showing the forest edge. Figure 5.2 Profiles of urinary pregnandiol in ng/mg creatinine and perineal swelling for the adult female WL during April 2007.2 Illustration showing how peak frequency of the fundamental and of the call was measured at each call stage.5 Mean number of individuals in the audience in the presence/absence of a copulation call. Figure 5.4 Individual variation in copulation calling behaviour. 73 73 77 78 91 Figure 6. Figure 4.1 Time-frequency spectrogram of a female copulation call from Budongo Forest during maximum tumescence.1 Illustration of how spectral parameters were measured.3 Mean peak frequency at copulation call beginning (Hz) for each female when copulating with high and low-ranked adult males. Figure 4. concealed in a rucksack. rivers.2 Schematic representation of experimental set-up. 62 Figure 5. 93 .1 Relationship between the adult female and male sex ratio and the occurrence of female led infanticide in the Sonso community of wild chimpanzees.List of figures Figure 2. roads and finally the grid system. compartments.1 Photograph illustrating the speaker used to conduct playbacks.1 Map of Uganda with the location of Budongo Forest reserve indicated. Figure 3. Figure 4. Figure 2.3 Copulation calls and the effect of male rank. Figure 4.

Appendix figure 1 A tantrum scream produced by infant Karo of the Sonso community. Uganda.3 Boxplots illustrating the total duration spent looking at the speaker and the total number of looks one minute after playback in response to three types of playback stimuli. 97 98 148 .Figure 6. Budongo Forest. illustrating how the four temporal and spectral parameters were measured from each scream condition. Figure 6.4 Boxplots illustrating the latency to look in response to playbacks in each condition.

I propose that vocalisations may represent important behavioural counterstrategies. for a long time. particularly when resources come under pressure. Taken together. females were simply considered passive pawns of male social manoeuvrings. My results regarding the production and acoustic structure of copulation calls suggests that these vocalisations play a crucial role in the lives of female chimpanzees. These observations are especially surprising because. I examined two commonly produced female vocalisations: copulation calls and victim screams from chimpanzees (Pan troglodytes schweinfurthii) in the Budongo Forest. . Uganda. In this thesis I address this by focusing on female competition in wild chimpanzees and the importance of vocal counter-strategies. chimpanzee females commonly produce victim screams and these calls have been shown to vary systematically with the severity of aggression experienced. A playback experiment showed that victim screams are meaningful to females and that listeners do not just respond to the acoustically most salient signals in their environment. exactly how females manage these social pressures is unclear. dissipating the risks associated with female competition. like their male counterparts. While we are beginning to understand the complexities surrounding female chimpanzee group living. female chimpanzees can be competitive and aggressive. enabling females to navigate successfully through their socially intricate world.Abstract A growing body of behavioural data is beginning to show that. During aggression. Females may use this information to keep track of out-of-sight agonistic interactions and make appropriate social decisions regarding whether to avoid an ensuing attack.

Chapter 1: Introduction Chapter One: General Introduction Summary In this first chapter I review what we know about. I conclude by hypothesising that females may use vocalisations in strategic ways to operate successfully through their socially intricate environment. S. I argue that female competition plays a significant. and address the gaps in our understanding of. I focus particularly on copulation calls and scream vocalisations from both the production and comprehension perspective.W. potentially dangerous role in the day-to-day lives of female chimpanzees and begin to assess how vocalisations may help females manage this little studied social pressure. Parts of this chapter have been accepted for publication in the following review article: Townsend. Finally. K. female competition in primates. Audience effects in chimpanzee copulation calls. Communicative and Integrative Biology 1(2). 1 . (accepted). & Zuberbuhler.

Since this discovery. correlating with the individuals life-long rank in the female hierarchy. females were generally considered peaceful. Kahlenberg. and solely preoccupied with raising their offspring. because “all females breed” (Wrangham.Chapter 1: Introduction Introduction The myth of the passive female has plagued primate behaviour studies for decades. In essence. and the importance of competition in the daily lives of female chimpanzees is still in its infancy in comparison to other primate species. However. with particular focus on the ways in which females compete and the outcome of such competition. 2006). primarily from a vocal perspective. 1997). males can always increase their reproductive success through extra matings (Trivers. In comparison to their male counterparts. The notion that females compete with one another was consistently ignored. Studying the vocal systems of non-human primates can give us vital insights into the intricacies of their social lives. 1972. less gregarious. the resources they compete over. allowing us to understand the importance of social processes that would otherwise be impossible to identify (Cheney and Seyfarth. 2007). In this thesis. Combining these 2 . I intend to address this issue. femalefemale competition in non-human primates has become a major research area. Our understanding of the extent to which chimpanzees compete in the wild. higher-ranking females do better (Hrdy and Williams. 1983). cracks in this assumption began to appear with the observations that certain female monkey species show highly variable levels of breeding success. In contrast. I look at female-female competition in a community of wild chimpanzees (Pan troglodytes scweinfurthii). The bias towards studying male competition was bolstered by the assumption that all female primates are uniformly successful at reproduction.

as this is often more easily quantifiable (Wrangham.and potentially inter-group competition for resources among females. The majority of the non-human primate literature to date has focused on competition for food. 1980) and understanding the evolutionary processes driving taxonomic variation in these relationships. which ultimately results in intra. and what strategies they may employ to alleviate such pressures. it is highly likely that females may compete for other resources. as soon as access to resources begins to limit female 3 . Koenig. 2002). 1997). These socio-ecological models predominantly focused on the effects of group size on food acquisition and the role of diet on home-range size. 1979.Chapter 1: Introduction two areas of study may elucidate what kinds of pressures female chimpanzees are exposed to. However. 1997). Nevertheless. However.. The most popular theoretical model for explaining the evolution of female social relationships and the network of behavioural interactions primarily invokes ecological factors (Wrangham. 1977. Scramble or contest? Competition will not generally occur if resources are so abundant that population growth is not limited (Sterck. 1987). emerging work on female-bonded primates shifted the focus to female-female social relationships (Wrangham. group size and population density (Clutton Brock and Harvey. This socio-ecological framework infers that the pressures associated with predation risk necessitate gregariousness in diurnal non-human primates. Background Origins of female competition and its effects on female social relationships The relationship between ecology and social organisation in non-human primates has been studied for over thirty years. Sterck et al. 1980. such as good quality nesting sites and potentially male mating partners.

highly dispersed or in very large patches relative to the size of the group. In these instances. 1989. the potential for aggressive exclusion of resources is likely to select for well-defined female dominance relationships. These relationships can be maintained by long-term alliances between related females (Chapais. For example. 1980). 2002. contest competition does involve an agonistic component and the strength of this competition is thought to explain variation in female social relationships. Scramble competition predominates when limiting resources are either of a low value. 1988. 1988. Kahlenberg et al. In contrast. This need for strong female alliances is thought to be one reason why many female primates remain in their natal community groups whilst males transfer (Sterck et al.. Sterck et al. Janson and Van Schaik. 1997). obtaining more of the resource themselves and thereby increasing their fitness levels (Wrangham. When scramble competition is the default competitive mode.Chapter 1: Introduction reproductive output. Contest competition on the other hand occurs when resources are monopolisable and. 1992) and also mutualistic coalitions with non-relatives... 4 . The nature and severity of these competitive social relationships depends on the overall abundance and distribution of resources within a habitat (Van Schaik and Van Noordwijk. certain individuals can systematically exclude others from the limiting resource. 1991). no overt agonistic interactions over access to resources are expected: all individuals within the group obtain roughly equal shares. in press a). Koenig. individuals lose access to resources as other conspecifics have already found and used them. competition is expected. often with stable linear hierarchies (Van Schaik. 1997.. due to competitive superiority. Isbell et al.

Kanyawara. Guinea: Bossou. Ellis. Uganda: Budongo. Creel 2005). and is maintained by alliances among female kin (Emery Thompson et al.. 1987. 1971. Kummer. Goodall. Females 5 . male chimpanzees are philopatric. our knowledge of the nature of competition between females is limited. Bercovitch and Strum. Chimpanzee social structure and ecology Chimpanzees form large. These can range in size from 20 to 160 individuals (Sugiyama and Koman. Boesch and Boesch Achermann. Henceforth. 2000.Chapter 1: Introduction Empirical evidence for female contest competition now exists for numerous primate species: dominants consistently outperform subordinates when resources become scarce (Gouzoules et al.. Harcourt. residing in their natal communities throughout their lives. 1986. 1982. 2005). 2007). immigrating into new communities when sexually mature (Goodall. Ngogo: Muller and Mitani. 1987. 2003. 1968. 1979b) and are distributed across Africa’s equatorial rainforests and woodlands. 1995. Nishida and Hiraiwa-Hasegawa. 2005). In contrast to most cercopithecine monkey species. Despite this growing evidence for female competition. This competitive regime is associated with female-aggression and philopatry. most notably the chimpanzee (Wrangham. fluid social groups called “communities”. chimpanzees associate on a fission-fusion basis (Nishida. Despite amassing almost 180 years of research on chimpanzee behaviour and ecology from 7 different field sites (Cote d’Ivoire: Tai. Females are the dispersing sex. 1993. 2005). Tanzania: Mahale. Reynolds. there have been a few female primate holdouts. 1986) where larger groups split into smaller “parties” which can vary in composition and size (Reynolds. Gombe. Altmann and Alberts. Within these communities. I will therefore focus on female competition in chimpanzees. beginning with a brief overview of the study species and its general socio-ecology. 1997).

1980. 1986. Their diet may be sporadically supplemented with meat obtained from hunting.. Chimpanzees are primarily frugivorous (Wrangham et al. and are often highly aggressive or even lethal (Goodall. although the frequency at which hunting occurs differs substantially between communities (Guinea Bossou: 5 hunts in 8 years (Sugiyama and Koman. Inter-community encounters between males may occur during patrols. agonistic interactions and alliance formations (Goodall. Emery Thompson and Wrangham. 2004. 2006) Male chimpanzees are considered more gregarious than females. 1998. Kahlenberg et al. although the degree of female gregariousness does differ between populations (Wrangham et al.. 1993). 1997. Tai chimpanzees: 267 hunts in 11 years (Boesch and Boesch Achermann. 2005) or alone with their dependent offspring (Goodall.. Wrangham. and because of the potential reproductive and feeding benefits associated with high rank. 2005). 2000).. 1986).. Kahlenberg. Emery Thompson et al. in press a). males spend considerable time maintaining or striving for higher rank. 1996. Non-oestrus females spend the majority of their time either in same-sex “mothering” parties (Reynolds. meaning they patrol a large home range incorporating the smaller core areas of multiple females (Chapman and Wrangham. Boesch. This can be done through grooming. but they are known to also consume young leaves and terrestrial herb vegetation (Boesch and Boesch Achermann. dominance displays. Doran. linear dominance hierarchy among the males of any one community. Pepper et al. 2000). 1999. Chimpanzees are known to use tools to obtain various nutritious food types such as 6 . Bates.Chapter 1: Introduction range within unevenly dispersed individual core areas (Wrangham and Smuts. 2007. Newton-Fisher. 2001. 1992.. Williams et al. 2003).Wilson et al. Wilson and Wrangham. There is a clear. 1999a. Males are territorial. 1986). in press.. 2000. 1987).

2000. 1989. Hannah and McGrew. Furthermore. Nishida. Wrangham et al. 7 . Whiten et al. honey (Nishida and Hiraiwa. Murray et al. in press b). 1997. Boesch and Boesch. McGrew. one study on the Kanyawara community of chimpanzees in Uganda recorded no aggressive interactions between females and only a small number of pant grunts in over 680 hours of observations (Muller. 2007). but this occurs seemingly in the absence of female-female aggression. 2003) dominance interactions are generally rare. in over 22 years of data collected by the Gombe study. 1979a. These fitness advantages for high-ranking females suggest female-female competition for resources has a significant contest component in chimpanzees (Kahlenberg et al. Females appear to show high variation in fitness. 1999. 1992. 2002). do not match those predicted by socio-ecological models. In fact. Emery Thompson et al.Chapter 1: Introduction nuts (Sugiyama and Koman. The social relationships between female chimpanzees.. Kahlenberg et al.. some female dyads were not seen to have a single dominance interaction (Pusey et al. Several studies on Gombe chimpanzees show that dominant females reproduced more quickly. Female-female competition in wild chimpanzees The subject of competition in female chimpanzees is not well understood.. higher-quality diets in comparison to low-ranked females (Greengrass. 1984. 1974). philopatry or alliances.. 2006. however. high-ranking females have been shown to spend less time foraging and have narrower. whilst some females are known to dominate others (Goodall.. in press a). As with hunting behaviour. 1973.. 1964. Nishida. due to the social organisation of chimpanzee communities. Wittig and Boesch. had higher infant survival and more rapidly maturing daughters than subordinates (Pusey et al. 1982) and insects (Goodall. the occurrence and frequency of such tool use behaviour varies across chimpanzee study sites and these observations have lead to tool use practices being described as cultural traits (Whiten et al. 1997). Similarly. Secondly. 1987). 2001). 2005. Firstly. 1986. Fawcett....

this was the first systematic evidence for direct intra-group female competition in chimpanzees.. Females were found to form distinct “core areas”. with certain core areas having greater access to preferred fruit trees than others (Emery Thompson et al.. 8 . instead implying that it is an important component of chimpanzee social lives. Recent observations of female competition Recent work from a number of chimpanzee communities across Africa suggest it is erroneous to dismiss female competition. adult females are not surrounded by kin (Goodall. specifically core areas containing a higher density of preferred foods (Kahlenberg et al. It is these observations of the lack of female sociality that have caused some to question the relevance and importance of female competition and dominance for chimpanzees (de Waal. as indexed by inter-birth intervals and infant survivorship. 1997) and this has been linked to differential use of the community home range. focusing on how females manage to occupy preferred areas. 2007).. 1986) and do not generally form alliances in their new communities (Emery Thompson et al. which are. Whilst such competition had been previously hypothesised (Pusey et al. Females residing in the Kanyawara chimpanzee community in Uganda show similar variation in fitness and rank as identified at Gombe (Pusey et al. Sterck et al. Their results suggest that Kanyawara females do indeed compete directly over long-term access to resources. Kahlenberg et al.. 1994. Females occupying the richer core areas subsequently demonstrated heightened ovarian hormone levels and hence improved reproductive success.Chapter 1: Introduction whereby males remain in their natal communities and females transfer at adolescence. 1997). in this case. 1982... 2007). 1997).. in press a). (in press a) further addressed this relationship. Baker and Smuts.

2004). only resources such as food patches. Resources in competition? So far. 9 . Kahlenberg et al. increases in female density have also been shown to negatively correlate with reproductive success (Williams et al. Therefore. or from putting heightened pressure on pre-existing limited resource supplies. That is. Whilst this increase in aggression was primarily due to high-ranking residents directing aggression towards immigrants.. resident females could gain significant fitness advantages (Kahlenberg et al.. 1993) and exactly how the competitive interactions between females affects access to this resource remains unknown (Breiehaen and Slagsvold. by aggressively preventing new females from accessing resources that may overlap with their own.. The importance of female-female competition for males remains “a vast area of ignorance” (Berglund et al.. the majority of research addressing female competition within primate species and other animal taxa focuses on resources considered to be related directly to the female’s or its offspring’s own survival. (in press a) hypothesise that these socio-ecological conditions intensify competition as it essentially presents a challenge to the resource access of resident females occupying the attractive core areas within these regions.Chapter 1: Introduction This contest competition for resources in Kanyawara occurred under specific socio-ecological conditions. rates of aggression between females were almost four times higher than in the absence of such immigration. and most obviously when multiple immigrants were present. Competition was particularly intense in the months following the immigration of new females into the community. During these times. At Gombe. in press a). nest sites and safety from predators (Palmobit et al. often of equal severity.. there were also simultaneous increases in aggression between longer-term residents. 1988). 2001).

also forming one-male breeding units. elegans). which subsequently leads to diminished likelihood of fertilisation... Among chimpanzees. Such promiscuous mating is thought to reduce sperm quality and quantity. such as their sperm for example. such that in some cases there can be as many as three adult females for one adult male (see chapter three). [Van Voorhies.g. and non-human primates are no exception. males copulate on average four times a day (Short.. 1989). nematodes. such that there are more females for each breeding male (Vincent et al. Dunbar and Sharman (1983) reported an inverse correlation between birth rates and the number of breeding females per male. the fact that females immigrate means that during periods of high-immigration (Kahlenberg et al.. 1981) and physiological testing in captive chimpanzees has shown that sperm quantity systematically decreases following consecutive copulations (Marson et al. (C.. For example. in gelada baboons (Theropithecus gelada) forming one-male breeding units. 1994]). 1992). (Lepidoptera Pyralidae). Similar sperm limitation has also been suggested in other animal species (e. multifemale promiscuous mating systems. [Gage and Cook. 1992]. sperm depletion resulting in female-female competition for mating partners is theoretically possible (Matsumoto-Oda et al. the resources males can offer. 1994). dominant rams (Ovis aries) have been shown to have reduced paternity success due to multiple matings with different females (Preston et al. 2001). 10 . moths. Recent research in species across the animal kingdom has shown that in low-male.. whilst in captive hamadryas baboons (Papio Hamadryas). in press b) the sex-ratio is female-skewed.Chapter 1: Introduction Females are expected to compete for mating access when either the sex ratio is heavily skewed. can become a limiting resource. For example. or if males vary in quality and their ability to provide a limiting resource. Firstly. 2007). the number of females in oestrus at any one time negatively correlates with the probability of conception (Zinner et al. Secondly.

1983) or preventing access to mates (Frame et al. Females may also compete aggressively not only over male sperm. Rood.Chapter 1: Introduction Given that females can cycle simultaneously (Goodall. Palombit 2000). 2004) access to male partners may therefore be restricted at times. males themselves may be an important resource that can be competed over. These socio-physiological factors suggest that sperm in chimpanzee mating systems can become limited. In chacma baboons (Papio ursinus) for example. In chimpanzees. females are contending for social bonds with males that may potentially result in some form of post-natal service for the female involved (Palombit et al.. 1980. as in other animal species. 1997. 2008). In the former case. in addition to their sperm. 2005. Stumpf and Boesch. 2003.. due to the “role-reversed” sex ratio. 1993. 1986. Berglund et al. potentially aggressively (Berglund et al. 1979. female chimpanzees may compete over this resource.... Deschner et al. though strategies may involve intense physical harassment (Wasser and Barash. but in what context could males provide a postnatal service or support? Competition for male support Intra-individual variation in lifetime reproductive success can be explained primarily by offspring mortality (Clutton-Brock et al. Therefore. Exactly what form the aggression can take is probably dependent on the strength of the competition occurring. 2008). Rosvall. 2001).. Competition for males differs both proximately and functionally from competition among cycling females for access to sperm. but also over the males themselves and the potential “help” they can offer. Rosvall.. 1988) and therefore a male’s ability to offer 11 . females have been seen to compete directly for access to particular male “friends” (Palombit et al. 1993. 2001.

. High-ranking males provide a vital peacekeeping service during intra-community male-female interactions (Goodall.. see chapter three). during female-female aggression (Pusey et al. Given the potential fitness benefits. Infanticide is a pervasive feature of most primate societies. 1985. 2000). 1984. Kahlenberg et al. males can enhance offspring survivorship by provisioning food or incubating eggs over gestation (Clutton-Brock.. Specifically. in press. 1994). 1997) benefits the perpetrator as he increases the chance to sire his own offspring and hence improves his reproductive success. 1991). 1986. including chimpanzees (Hausfater and Hrdy. This mostly male act (Tuomi et al. Pusey and Packer. protection against infanticide may be one attractive service provided by community males. in press b).Chapter 1: Introduction offspring protection is likely to be a very attractive trait for females (Palombit et al. 12 . males may be capable of providing support against infanticide.. Male chimpanzees therefore provide at least two important forms of post-natal service: protection against infanticide and day-to-day protection against male and female aggression. as recent research shows. Infanticide appears to be a sexually selected strategy that accelerates the return to fertility in lactating females (Struhsaker and Leland. Therefore. Goodall. In birds. and.. 2001). 2001). however in mammals alternative forms of post-natal investment predominate (Palombit et al. Boesch and Boesch Achermann. 1986. Chimpanzee males may also be valuable to females for the support they offer during more general agonistic interactions. it is plausible that female chimpanzees compete with one another for access to protector males.

in press b). (in press b) suggest that one possible counter-strategy to dissipate the aggression that parallels female competition is to take advantage of male protective services. East et al. in press.Chapter 1: Introduction Counter-strategies against female competition It is now clear that female competition. It is therefore highly likely that selection would favour the evolution of counter-strategies available to less competitively able females. 13 . 2004) and lions (Panthera leo) and hyenas (Crocuta crocuta) (Packer et al. From this. remained unclear. some female primates are known to mate promiscuously with multiple males in order to confuse paternity (Hrdy. 1979.. For example.. immigrants associate more with adult males and sometimes with other immigrant females. Research from different mammal species indicates that behavioural counter-strategies are often employed to avoid conflict associated with group living. These systematic results mirror more anecdotal findings at other research sites that indicate females seek out the protection of males. 1989) both suggest that nulliparous females spend more time with males in comparison to other age-sex classes. exactly how female chimpanzees manage the risks associated with intra-sexual competition has. Kahlenberg et al. Their analyses have shown that in the face of heightened female aggression.. Van Schaik et al.. (in press a... and the potential aggression that can accompany it. 1980) and Mahale (Nishida. Kahlenberg et al. Kahlenberg et al. such as low-ranking or immigrant individuals (Pusey et al. is a considerable risk to female chimpanzees. Observations at Gombe (Pusey. However.b) addressed this question specifically by analysing almost 11 years of long-term data from the Kanyawara chimpanzee community. 1989) have been shown to avoid the core of the group around parturition – both behavioural counter-strategies against infanticide. until recently. 2001..

may have evolved as a cheap form of “social grooming”. 2003) and stimulating travel (Rendall et al. preventing larger social groups from breaking down. and thus maintain social bonds. Gouzoules et al. Vocalisations have even been highlighted with facilitating the evolution of complex sociality. This is particularly surprising given the vital insights vocalisations have given us into the social intricacies of other primate species (Cheney and Seyfarth. and consequently could break down. For example. mediating reconciliation (Wittig et al. 14 . inciting aggression.. Vocalisations however would allow contact with other group members in the absence of mechanical interaction.. 1977. 2003) argues that grooming in primates is essential to maintain social bonds and group cohesion.Chapter 1: Introduction The role of vocalisations Chimpanzees can manage the pressures of female competition using behavioural counterstrategies and vocalisations are likely to play a key role in this. 2005). 2007. therefore. Goodall. McComb and Semple (2005) have provided further evidence for this hypothesis. challenging dominance status (Kitchen et al. Their phylogenetic meta-analyses of primate vocal repertoires and grooming time (their measure of sociality) show a clear correlation between vocal and social complexity.. the dense nature of the forest habitat and the fission fusion nature of the chimpanzee social system mean that vocalisations are probably the most important communicative tool for chimpanzees. Despite their importance and ubiquity. Dunbar (1993. being produced in almost every social context (Marler and Tenaza. 1986).. Due to time constraints some social relationships could not be continually serviced through grooming. Furthermore. 1990. Complex vocal systems. 2007). Whilst they point out the direction of causality cannot be inferred from correlation analyses. vocalisations represent important social tools. 1986). we still know very little regarding the functions vocalisations play in the daily lives of chimpanzees (Slocombe. 2000). in baboons. Vocalisations are possibly the most pervasive aspect of chimpanzee life.

(Fischer et al. 1977). 1991). including chimpanzees. Since then. Procolobus badius (Marler. have been shown to possess a graded vocal repertoire (Red Colobus monkeys. I will ask the questions: how extreme is femalefemale competition and how do females use vocalisations to manage the risks and pressures associated with such competition? This research will give us a better idea regarding the complexities of female social life and the potential strategies they employ to navigate successfully through their complex social worlds. (Marler. numerous other primates. 1970). these calls can be used as objective data concerning their “thoughts” and motivations (Griffin. focusing on vocalisations could open a previously closed window into the mind of the competitive female chimpanzee. in terms of their acoustic structure. Graded vocal systems differ from discrete systems primarily in the amount of information they can potentially encode. According to Marler (1976). Hence. Rhesus macaques were one of the first primate species shown to utilise a graded system with grading between and variation within call types (Rowell and Hinde. Papio cynocephalus ursinus.Chapter 1: Introduction their results are consistent with the hypothesis that the vocal communication system and social complexity are intimately related. Chimpanzees Pan troglodytes schweinfurthii. Vocalisations also represent an important medium from which to assess the underlying cognitive sophistication of animals.. if there is a highly ordered relationship 15 . When animals vocalise. Baboons. In this thesis I aim to address these issues in chimpanzees and make progress towards filling the gaps in our knowledge. 2001). 1976. 1962). The chimpanzee vocal system Animal vocal systems can vary on a continuum from discretely organised to highly graded. Specifically. Marler and Tenaza.

More objective call categorisation employs mathematical representations of acoustic distinctiveness.1 for list of call types) Table 1.Chapter 1: Introduction between signal grading and context of production. a number which was then doubled by Goodall (1968) who isolated 24 specific call variants. feeding. graded systems can potentially transmit vast quantities of information and hence the scope for vocal complexity is much greater. identifying over a dozen different call variants – a finding which has recently been confirmed by Slocombe (2005). The repertoire Reynolds (1965) identified roughly 12 different call types. the extremely graded nature of the chimpanzee vocal system has made it very difficult to conduct systematic investigations and to construct a clear vocal repertoire. which is subsequently put in relation with the context of production (Slocombe. Marler and Tenaza (1977) undertook this rather significant challenge. but lacked a quantitative basis.1: The 14 different chimpanzee call variants and their contexts of production (adapted from Marler and Tenaza. SOS screams) Aggression 16 . 2005). These estimations were primarily based on how the call sounded and the context in which it was produced. Call type Pant hoot Pant grunt Laughter Squeak Whimper Barks Waa bark Cough Rough grunt Pant Grunt Wraa Huu Context Travel. However. resting Greeting Play Copulation Conflict/begging Social excitement/aggression Danger Stop undesirable behaviour Food Greeting Resting Not clear Unusual events Scream (incl. 1977 and Slocombe. 2005). suggesting this final estimation is robust (see table 1.

2). same-sex parties or alone. Because of their ubiquitous occurrence and often complex acoustic structure. with much debate surrounding the proximate and ultimate functions of this conspicuous behaviour. much research has focused on these vocalisations in the primate taxa (Hauser. 1978. Explanations for their pervasiveness range from calls just being incidental byproducts of the copulatory act. In fact. in many populations females spend the majority of their time in small. 1986. Due to their less social nature.Chapter 1: Introduction Female vocalisations Female chimpanzees. Therefore. to calls synchronising orgasms between the mating partners (Hamilton and Arrowood. see table 1. only accompanied by their dependent offspring. female chimpanzees do not vocalise at rates comparable to males (Goodall. These so called “copulation calls” have intrigued generations of evolutionary biologists. are less gregarious than adult males. Slocombe. mating events are accompanied by vocal signals. on the whole. 2006 for more details). 1996). Copulation calls are defined as acoustically distinct vocalisations produced prior to. there is a restriction over which call types could be investigated in this thesis. As previously described. I primarily focus on the more common vocalisations produced during evolutionary important events: the female-specific and reproductively important copulation call. and the socially valuable victim scream produced during aggression. to date 13 different functional hypotheses have been put forward (see Pradhan et al.. 1998b). all of which 17 . As the thesis progresses it will become apparent what role each of these two vocalisations has in helping females manage social pressures. unpublished data) and as a consequence. Female copulation calls Across the animal kingdom. during or just after copulations (Semple.

2: The functional hypotheses so far presented to explain the adaptive function of primate copulation calls (adapted from Pradhan et al. This is 18 . 2006). Hypothesis Function Reference 1 Orgasm-like reaction and non-adaptive byproduct of the copulation Hamilton and Arrowood.. One of the most commonly invoked explanations suggests that these acoustic traits function to alert males to the receptive condition of the female. 1996 3 Self stimulate ovulation Cheng. 1990. subsequently inciting competition amongst the males (Hauser. Ultimately the female benefits by getting the best sperm from the best male. Only one study to date has assessed these ubiquitous vocalisations in any ape species (Hauser. 1992 4 Synchronize orgasm Hamilton and Arrowood. 1978 2 Non-adaptive maintained by phylogenetic inertia Henzi. Henzi. 2002). Despite the wide range of explanations. 2001. however. 1978 5 Strengthen the pair bond Hamilton and Arrowood... 1996 13 Promote mate guarding by preferred mating partners Maestripieri and Roney. 1990). 2005 The majority of the 13 hypotheses put forward to explain the adaptive significance of primate copulation calls. 1994 11 Incite paternity confusion and protection from infanticide O'Connell and Cowlishaw. O’Connell and Cowlishaw. 1994 12 Announce paternity certainty to promote paternal investment Henzi. such that the female ends up with the most dominant mating partner. Semple et al. 1977 9 Incite male-male competition Cox and Le Boeuf. 1990 7 Advertise sexual motivation Gouzoules et al. have originated from research on monkey species. or at a smaller though no less important scale. Table 1. between sperm. 1978 6 Advertise fertility Aich. Competition can occur directly between males.Chapter 1: Introduction share a common theme: copulation calls are sexually selected traits that benefit the calling female’s reproductive success. Hauser. 1996. 1998 8 Promote breeding synchrony between females Viljoen. some traditional hypotheses still seem preferred over others. Semple. 1994. 1990 10 Incite sperm competition O'Connell and Cowlishaw. 1977.

Furthermore. Kahlenberg. it is entirely plausible that female competition may have had an additional impact on the evolution of this reproductively important vocal behaviour. at some level. in press. including chimpanzees (Koenig. As competition can be detrimental to females.. it would also be valuable to elucidate what additional information is encoded within these calls. From a total of 64 copulations observed over one month. females were more likely to receive aggression following vocal copulations with young adult males than established adult males. in press a). male-male competition is only one aspect of a large spectrum of social processes that take place in the reproductive context. female-female competitive interactions over resources (including food and males) and the aggression that accompanies it are now known to be a pervasive and threatening feature of non-human primates. Hauser found females called more with “established” adult males than “young adult” males.Chapter 1: Introduction surprising given the crucial role that apes can play in understanding the adaptive significance and evolution of human behaviour. Hauser loosely argues that. 2006). and copulation calls may play some role in maximising an individual female’s fitness. 19 . 1997. although this relationship was not significant. Yet. from a phylogenetic perspective (Tomasello. The results of Hauser’s (1990) brief study on chimpanzee copulation calls are inconclusive. female copulation calls may operate to incite competition amongst males. As previously outlined. just as they are thought to in other primate species. Female chimpanzees therefore play a more influential role in the lives of other females than was formerly assumed (Pusey et al. However. To address this question. one avenue involves focusing purely on the contexts of copulation call production. as is normally done in studies of this type. 2002.. Kahlenberg et al. in press).

Semple et al. 2005).... For instance. Semple et al. 2007). In this detailed analysis of calling behaviour. (Deschner et al... how fertile she was. this finding was inferred solely from the size of the sexual swelling (Semple et al. 1977).Chapter 1: Introduction Information content Research on numerous animal taxa has shown vocalisations can provide listeners with a variety of information sets. 2002). their maturity. Subsequent playbacks of these acoustically distinct ejaculatory and non-ejaculatory calls to male subjects suggest that the information content of these calls is indeed salient to listening males. Calls can also transmit information regarding the effective state of the caller (Morton.. 2004. e. 2003) and hence this result should be viewed cautiously. However. sexual state and condition (Clutton-Brock and Albon. (2008a) controlled for numerous confounding variables such as male rank and proximity to ovulation. barbary macaque copulation calls have been shown to acoustically vary primarily with occurrence of ejaculation (Pfefferle et al. This time ovulation was systematically elucidated from fecal and urinary hormone analyses (Deschner et al. calls can allude to the identity and size of the caller. Heistermann et al..g. (2002) also found that calls varied with the female’s proximity to ovulation. Reby et al. 2008a). 2002). It has now been shown that the temporal relation between swelling size and proximity to ovulation (when the female is fertile) is not an accurate indicator of reproductive state in primates. For primate copulation calls the following studies have been conducted: In yellow baboons (Papio cynocephalus ursinus) copulation calls have been shown to vary systematically with the rank of the male mating partner and her identity (Semple. More recently.. 1979. but was found to have no explanatory effect. 1977) and their subsequent behaviour (Smith. 2001. potentially 20 . Pfefferle et al. 2003..

chimpanzee pant hoots have been suggested to convey information regarding the behavioural state of the caller (e. They dismissed the potential for information transfer and alternatively focussed on the physical mechanisms driving the acoustic variation in the calls. 2005b. mating vocalisations. Crockford and Boesch (2003) examined the circumstances of bark production in adult males of the Tai forest. 1996). 2005b). their work has shown that the acoustic structure of food grunts varies systematically with the quality of food eaten (Slocombe and Zuberbuhler. or indeed ape.Chapter 1: Introduction allowing them to make behavioural decisions about when to next copulate (Pfefferle et al. 2006). and found clear acoustic subtypes were produced in response to snakes. despite advances in our understanding of the potential informational content of other call types. They proposed that the height above the ground and energetic output of the caller might better explain the acoustic differences observed in this call type. travelling or feeding) and information about the environment (e. These acoustic differences were also shown to be meaningful to receivers at some level. Crockford and Boesch suggest barks may be capable of informing listeners about external events such as a potential danger. From this. Slocombe and Zuberbuhler (2005a. or the occurrence of a hunt.. and while hunting. No study so far has addressed the acoustic structure of any chimpanzee. With regards to foodassociated calls. Chimpanzee barks might also convey discrete information sets to listening individuals. A comparative wild study on scream vocalisations 21 . 2005b. Most recently. Ivory Coast. A similar study on these long distance vocalisations performed by Notman and Rendall (2005) gave identical results. 2008b.g. presence of a large food source) (Uhlenbroek. 2006) have focused on the food grunts and screams of chimpanzees. although they chose to interpret their findings differently.g. suggesting chimpanzees could extract information regarding food type encountered (Slocombe and Zuberbuhler. For example. see also Semple and McComb. 2000).

of these vocalisations in the lives of female chimpanzees (Semple et al. Victim screams Agonistic interactions are a natural outcome of social living and primates can produce acoustically distinctive vocal signals during these events. 2005a.. Exactly what information is conveyed by these calls has been an area of particular interest in monkey species and more recently in chimpanzees.Chapter 1: Introduction indicated that the role an individual plays in fights and the severity of aggression experienced by the victim might also be encoded within screams (Slocombe and Zuberbuhler. Male and female chimpanzees produce 22 . we are likely to gain a better understanding regarding the function. Recent research has shown that chimpanzee screams produced during agonistic interactions also have the potential to convey social information. 1984). Rhesus macaques (Macaca mulatta) have been shown to produce acoustically distinct scream variants as a function of aggression severity and the relatedness and relative rank of the opponent (Gouzoules et al. 2007). 1984). 2002). or role. conveying information regarding the nature of the interaction in progression to potential listeners. such agonistic vocal signals may confer an adaptive function. In dense natural habitats where visual contextual information is limited. By supplementing the contexts of call production with what information copulation calls may also encode.. I focus in detail on these particular studies later.. Playback experiments of infant screams to mothers further indicated that these screams were indeed meaningful to recipients (as measured by looking duration) and that information regarding the ongoing agonistic interaction could be extracted from the acoustic structure of screams (Gouzoules et al.

1986. Nonetheless. female chimpanzees are equally capable of extreme.Chapter 1: Introduction acoustically distinct screams as a function of the social “role” they play in a fight. 2005. Wilson and Wrangham. Victim screams have also been shown to exhibit considerable variation in their fine acoustic structure. produced more often. 2000. Wilson et al. aggression has traditionally been looked at from the male perspective. with recent reports of coalition formations amongst females (NewtonFisher. and were longer in length than those elicited during mild aggression (chases or bluff displays). in press b) and attempted female-led infanticides (Pusey et 23 . before this conclusion can be confirmed. but when they did. similarly to rhesus macaques. 1993. In chimpanzees.. albeit less commonly. specifically whether they are the victim or the aggressor (Slocombe and Zuberbuhler. 2003. Kahlenberg et al. experimental studies are required to systematically show these calls are meaningful to chimpanzees at some level. 2001). 1990). intra-sexual aggression. Fawcett and Muhamuza. stampings) were higher in frequency. Chimpanzee victim screams therefore also have the potential to signal to listening individuals how severe the ongoing aggression is. Screams produced during severe aggression (beatings. From this. However. chimpanzee victim screams varied systematically with the aggression experienced (Slocombe and Zuberbuhler.. 2007). the screams they produced were acoustically very different to victim screams. 2005a). Cheney and Seyfarth. Slocombe and Zuberbuhler (2005a) postulate that the structural differences make these screams potential candidates for functionally referential signals – signals that refer to objects or events in the external environment (Macedonia and Evans. primarily due to its more frequent occurrence and sometimes-lethal nature (Goodall. Aggressors seldom vocalise during agonistic interactions in comparison to victims. More detailed analyses elucidated that.

Most importantly. and the likelihood of the aggression escalating or snowballing towards them. Fawcett 2000). Secondly. is most apparent when resources become scarce. 1997. in press. it seems. 1986. a playback experiment must be employed. playback experiments have made 24 . This would be particularly important in communities where the female offspring of high-ranking females remain resident in their natal community for life (Pusey et al.. 1990). see chapter three for more details). Such aggression. Whilst there are clear advantages for processing the information in victim screams. Reynolds. 1997. it is as yet unclear whether these screams are meaningful to female listeners..Chapter 1: Introduction al. These experiments represent a powerful technique for investigating the meaning and function of animal vocalisations. In order to systematically test this. 2000. Over the last 30 years. Playback experiments can be designed to reproduce naturally occurring events or to present subjects with a stimulus they normally would be unlikely to encounter (Cheney and Seyfarth. It is clear from these recent advances in understanding.a behaviour observed often late into the offspring’s life (Goodall. that being able to extract information regarding the severity of aggression occurring in the nearby vicinity would confer selective advantages for females. attending to female aggression severity also allows females to make informed decisions as to whether to intervene on behalf of their offspring. Understanding the nature of aggression may therefore play a significant role in the day-to-day lives of female chimpanzees. such comprehension would allow females to keep track of potentially lethal intra-sexual agonistic interactions. helping them dissipate the risks associated with female-competition and navigate successfully through a socially challenging world. Boesch and Boesch Achermann. 2005).

In chapter three I will assess how relevant-a-pressure female-female competition is. such as inciting male-male competition. Possibly due to the inherent difficulties associated with conducting playbacks in the wild (Cheney and Seyfarth.. 2000). no experiment has so far addressed ape call comprehension in a naturalistic setting. I hope to better understand the complexities of the social world in which female chimpanzees live. and possibly more pertinently. only the vocal system of chimpanzees has been experimentally probed in this respect. Slocombe and Zuberbuhler (2005b) used playbacks to address chimpanzee call comprehension. experimental and acoustic research techniques. whether traditional hypotheses. and how females use vocalisations to successfully manage these demands. 1990. conveying information regarding food quality. only a handful of studies have employed playback experiments with any ape species. Firstly.Chapter 1: Introduction vital contributions to understanding the communicative and cognitive skills of many bird and mammal species (Macedonia and Evans. reviewed in chapter six). in the lives of Budongo chimpanzees. Despite the success of playback experiments in furthering our understanding of animal vocalisations. Wilson et al. demonstrating that food-grunts have the potential to function as referential signals. (2001) and Herbinger (2004) used the long-distance pant-hoot vocalisations of chimpanzees to investigate intercommunity conflict in East and West African chimpanzees. Secondly. This experiment was conducted in a captive setting where experimental manipulations are often more manageable (Hare et al. In chapter four I will focus on the contexts in which female chimpanzees use copulation calls. In fact. are sufficient to explain call production and function and whether copulation calls are additionally affected by 25 . 1993). Outlook By combining observational.

I will also take into account what additional information these calls can encode in their fine acoustic structure. By designing and employing a customised playback paradigm I will experimentally test whether female chimpanzees can comprehend the severity of aggression from victim screams and will consider ultimately.Chapter 1: Introduction the social pressures accompanying female competition. what strategic role these vocalisations might play in the day-to-day lives of female chimpanzees. Together. In chapter six I will move away from copulation calls and focus on female understanding of victim screams. I hope such research will make vital steps towards overcoming the long-held myth of the passive female chimpanzee. 26 . but rather than purely focusing on the factors that influence call production. what this tells us about the importance of understanding scream vocalisations for female chimpanzees. In chapter five I will build on the findings of chapter four. crucially. chapters four and five will reveal the evolutionary pressures under which chimpanzee copulation calls have evolved and.

swamp forest and ironwood forest (Plumptre. A short dry season between December and February intersperses two rainy seasons.100 metres and has a mean annual rainfall of about 1. Due to a history of logging. The study site is located at an altitude of 1. The Budongo Forest lies between 31o8 and 31o42 East and 1o35 and 1o55 North.1: Map of Uganda with the location of Budongo Forest reserve indicated (courtesy of Slocombe. 2005). these are reviewed more extensively with each chapter. 2005). 2005). Uganda. colonizing forest. The forest covers an area 428km2 and is classified as moist. 2005).2).Chapter 2: Methods Chapter two: General methods This chapter will provide a description of the chimpanzee community studied.1 and 2. the forest comprises a combination of four forest types: mixed forest. semi-deciduous tropical forest (Eggeling. Because certain procedures are specific to each research topic. 27 . 1992) of the Budongo Forest Reserve. Monthly maximum and minimum temperatures vary between 32oC and 19oC (Reynolds. Study Site Data were collected over a 14-month study period between January 2006-November 2007 on a habituated research group of chimpanzees in the Sonso region (Reynolds. 1997). 1947). Figure 2. in the Masindi District of Uganda (figure 2.600 mm (Reynolds. March-May and July-November (Reynolds. the research site in Uganda and a very brief introduction to the more different data collection protocols used in this study.

compartments. Tanazania. the Sonso community. chimpanzees (Pan troglodytes schweinfurthii). Similar chimpanzees are found in the Kanyawara and Ngogo communities in Kibale National Park. 2005). roads and finally the grid system.Chapter 2: Methods Forest and compartment boundaries Roads and tracks Rivers Budongo Forest Project grid Figure 2. Uganda and in the Gombe and Mahale National Parks. Whilst there exist almost 650 chimpanzees in the entire Budongo Forest (Reynolds. centred at the core of the Sonso community’s home range (courtesy of www. or long-haired.budongo. only one community. Study community The chimpanzees in the Budongo Forest belong to the subspecies of eastern.2: Map of the Budongo forest showing the forest edge. is sufficiently 28 .

1: The adult and sub-adult males of the Sonso community present during the 14 month study period. Table 2.KE.Chapter 2: Methods habituated for human following and research. A systematic habituation program began at this time. Names in bold served as playback subjects. During this period. All but a few peripheral adult females were confident when in mixed parties. The Sonso community of chimpanzees are named after the river running through their home range and have been continuously monitored by researchers and field assistants since 1991. All individuals are named and given a two-letter code derived from letters of their name. same-sex parties.SH. under the supervision of Vernon Reynolds.KM Died March 2008 Orphaned August 2007 Disappeared August 2007 Pascal (PS) Disappeared April 2008 Simon (SM) 29 . and provisioning has never been used.JT. At the time of study (2006-2007) the community consisted of 78 individuals including 8 adult males and ~ 25 adult females (see table 2. permitting researchers to follow them.2). adult males were the most habituated.1 and 2. Adult males Nick (NK) Duane (DN Zefa (ZF) Bwoba (BB) Maani (MA) Musa (MS) Bob (BO) Gashom (GS) ~Age 26 42 25 21 50 17 18 21 KR Known offspring Details Alpha Sub-adult males Kato (KT) Kwezi (KZ) Hawa (HW) Zalu (ZL) Fred (FD) Mark (MR) ~Age 15 12 13 13 14 11 10 15 Details RS. even when travelling alone. but the majority remained difficult to follow when travelling just with offspring or in small.

Females in bold served as playback subjects (see chapter six). Asterisks denotes females that served as subjects in the copulation calling study (see chapters four and five).Chapter 2: Methods Table 2. their dependent offspring and adolescent females of the Sonso community during the study period.2: The adult females. Adult Female Biterice (BC) Banura (BN) Flora (FL) Gladys (GL) Harriet (HT) Juliet (JL)* Janie (JN) Age 32 40 29 32 30 18 24 JT JS BH KM KC Offspring BG BT FA FK GN HL HY Details Adolescent female Age 15 11 Anna (AN) Missing July 2007 Rachel (RE) Kalema (KL) 29 Kewaya (KY)* Kigere (KG) 24 32 KA KX KE KI KN KS KR KB Kutu (KU)* Kwera (KW) Lolla (LL)* Melissa (ML) Mukwano (MK)* 29 27 20 25 28 MN NR NT RM RS SL SN ST Alpha female Nambi (NB)* Ruhara (RH) Sabrina (SB) 46 40 27 Sarine (SE) Wilma (WL)* Zana (ZN) 37 27 27 SK ZD ZG ZK Died Aug 2007 Zimba (ZM) 40 30 .

51km2 (Newton-Fisher 2002b.budongo.3) and is surrounded by three communities of unknown size and Chimpanzees would be located in the forest though de-nesting previously nested chimpanzee groups or by listening for and following vocal cues. 31 . The main road indicated is called the “Royal Mile” and provides access to the study site (courtesy of www. Data collection Over the 14-month period I collected data on 260 days.78 and 14. Inter-community relations are in-frequent and there has been no recording of inter-group killings (Reynolds 2005).3: Map of the grid system used to navigate the Sonso community home range. Roads and tracks River Sonso Budongo Forest Project grid Figure 2.Chapter 2: Methods The home range habitually used by the community has been estimated as between 6. Each block covers an area of roughly 100m2 and is allocated a letter and a number to aide navigation through the forest. 2003 see figure 2. constituting approximately 2000 observation hours in the forest.

1974). longer whole day samples (6am-7pm) were taken over a period of 14 consecutive days. For copulation calling studies. 32 . then one was chosen at random following the protocol outlined in Newton-Fisher (1997). In addition ad-libitum data on chimpanzee pant-grunts were collected. If multiple focals were present. focal samples would last whole days (7am-4pm) and focal individuals were chosen based on a number of factors. The location is documented and then the same site is returned to early the next morning (approx 6am) before the focal leaves the nest. the focal had to be nested and de-nested. Pant-grunts are submissive vocalisations given by subordinates to dominants and are generally regarded as good indicators of dominance status (Goodall 1986). both focal and ad libitum animal sampling methods were employed. For playbacks.Chapter 2: Methods Sampling methods When conducting playback experiments and recording the behaviour of oestrus females. In theory the identity and the duration of focal periods should be specified prior to commencement. Focal sampling has been defined as when “attention is directed to one target individual at a time” (Altmann. Behavioural and acoustic data on copulation calling was supplemented with ad-libitum data. Frequency. This was primarily because I had to remain with the female for the duration of her receptive period (see chapter four for more details) and in order to do this. This was only taken if the individual copulating could be clearly seen and did not interfere with focal data sampling. This observational technique involves staying with the female until she constructs and rests in a nest. though in reality. due to the ranging patterns of wild chimpanzees. this is often not possible. particularly when he or she had received their last experimental trial (see chapter six). This is a particularly useful methodology as it allows all instances of specified behaviours exhibited by that one individual to be recorded.

Equipment To collect acoustic recordings of copulation calls and playback stimuli I used a SENNHEISER ME66 directional microphone and MARANTZ PMD660 solid-state recorder. or Fujitsu Siemens Amilo laptop computer. 33 . including scripts written by M.Chapter 2: Methods grunter ID and receiver ID would be noted and later entered into a pant-grunt matrix. Playbacks and infanticides were filmed using hand held video cameras (Panasonic NV-GS 250 and Canon MVX330i). This allowed systematic dominance relationships to be elucidated (see chapters four and five). File editing and quantitative call analyses were conducted with PRAAT software version 4. Owren.37. For more details on specific methodologies see chapter methods sections.3. Recordings were directly transferred as WAV files to a Toshiba Celeron. Playbacks were conducted using a NAGRA DSM speaker/amplifier. Calls to be broadcast were stored and played in uncompressed WAV format on an Ipod Nano. .

K. Recent research from wild chimpanzee communities suggests that such female aggressive behaviours may be a product of competition for resources. The results from this chapter have been published in the following paper: Townsend. M.Chapter 3: Lethal female competition Chapter Three: Lethal female competition in wild chimpanzees Summary Wild chimpanzees exhibit substantial gender differences in behaviour. As a consequence males can display extremely violent behaviour towards neighbours. and occasionally community members. Slocombe. I report here recent observations of lethal aggression in the form of infanticide perpetrated by resident females of the Sonso community. Emery-Thompson. other primates. & Zuberbühler.W. Uganda. (2007). Adult males typically dominate the females and compete amongst each other for access to resources. S. With these observations I directly test this hypothesis in addition to examining a number of other socio-ecological factors that bring out the hidden demonic nature of chimpanzee females. Female-led infanticide in wild chimpanzees. females are regarded as the more peaceful sex. One observed and two inferred cases demonstrate that gender differences in aggressive propensities in our closest living relatives may be much less pronounced than originally thought. who lead relatively secluded uncompetitive lives. Budongo Forest. predominately occupied with raising offspring.. Current Biology 17 (10): 355-356 34 .. K. In contrast.E.

1977.. 1994). are the selective advantages for males and females of this extreme form of aggression? For males. 1982. 24 cases of infanticide have been documented. competing with conspecifics to maximise mating success is likely to be the most powerful force driving this behaviour. 1986. Ngogo: Watts et al. Females were involved only four times. 1993. 1985. In 14 cases. in one case jointly with a male. Most functional explanations resort to sexual selection theory. 1997).Chapter 3: Lethal female competition Introduction Infanticidal behaviour has been observed in a number of mammalian species and is mostly a male affair (Tuomi et al. In East African chimpanzees. Goodall. primates in which males migrate and typically obtain high rank in the new group after arrival. occurring at each of the five long-term study sites (Budongo: Suzuki. Hamai et al. 1972. 1994). For example. demonstrating that males acted as aggressors in the majority of cases (N=11). Newton-Fisher.. 1977) and chacma baboons (Palombit et al.. 1999b. Pusey and Packer. 1971. What. 1985.. 1997). Takahata. 1985. 1979. if any. Bakuneeta et al. Kawanaka. infanticidal behaviour is far less uniform and it has been reported from both males and females (Hiraiwa-Hasegawa & Hasegawa. 1999. Nishida and Kawanaka. thereby increasing the chance to sire his own offspring (Struhsaker and Leland. 1981. 1992. suggesting that males obtain a direct reproductive benefit by shortening the victim female’s time to oestrus resumption. In chimpanzees. Gombe: Bygott.. observation conditions were sufficient to identify the individuals involved. 2002). Kanyawara: Clark-Arcadi and Wrangham. Norikoshi. Classic examples for this sort of behaviour come from langurs (Hrdy. it has been suggested that male infanticide in chimpanzees may be a counter-strategy to female promiscuous mating by coercing resident females into restrictive mating. a way to eliminate potential future 35 .. Mahale: Nishida et al.

Three of the four reported cases of female-involvement come from a mother and daughter dyad. which ultimately provide them with reproductive advantages (Williams. Clark-Arcadi and Wrangham. 2003). Nishida and Kawanaka. 1986). the empirical evidence is generally weak. Boesch and Boesch-Achermann. Takahata. 1999). Clark-Arcadi and Wrangham. explanations for this behaviour are even more difficult to formulate because their participation is so rare. though this is likely to be the case (Clark-Arcadi and Wrangham. More recent observations of intense aggression amongst high-ranking Gombe females. given their rarity and the relatedness of the dyad. These results provide a direct test of Pusey et 36 . including apparent attempts at infanticide. 1986). aggressors have not always had increased mating access to the mothers of the victims (Goodall. 1985. 1999. in addition to increasing future mating chances (Hamai et al. 1992. 1999) and male infants are not preferentially targeted by infanticidal males (Wilson and Wrangham.. led to the hypothesis that increased competition for resources can lead to violent behaviour amongst females (Pusey et al. Although these are plausible ideas. 1985. 1986). Emery Thompson et al. For females. 2000). In some occasions. infanticidal males have been identified as the father of the victim (Nishida and Kawanaka. either directly via meat consumption or indirectly via access to high-quality core areas.Chapter 3: Lethal female competition competitors. 1985. in review). Here. Takahata. I describe three new cases of female-led infanticide in wild chimpanzees. 1985. Finally. However. Passion and Pom at Gombe (Goodall. these infanticides have also been highlighted as potentially just pathological behaviours (Goodall. In all three instances the females ate the offspring and therefore nutrition has been mentioned as a crucial factor... 1999. There are also insufficient data that clearly demonstrate that inter-birth intervals are shortened following infanticide. in collaboration with Melissa Emery Thompson and Katie Slocombe. 1997).

ML. now joined by the first party of females (NB. consisting of a group of adult females (ZM. continuous screams could be heard from a second party to the South. are recorded on an ad libitum basis. The cases described here occurred between March 2004 and July 2006. WL). ML. and lost possession of the carcass to her. Uganda. were with a party of four adult females (KU. WL. when followed.Chapter 3: Lethal female competition al. Katie Slocombe and a field assistant. KL). all in a Cynometra alexandrii tree.’s hypothesis and ultimately suggest that females may have similar propensities for extreme physical violence as previously described for male chimpanzees. it was possible to film parts of the events using hand-held video camcorders (2004: Sony CCD-TR728E. blood was seen on the ground. BN. Raymond Ogen (KS and RO). alpha female NB with adolescent son MS). 2006: Canon MVX330i). 152) On 12th March 2004 at 13:50. She was charged by NB. The adult female ZM was first seen carrying a dead male infant. Three minutes later. KS and RO located the party. In two of them. They also saw an adult male (ZF) displaying in a tree. KU. Methods Observations were made as part of the ongoing behavioural research with the free-ranging Sonso community in Budongo Forest. whilst important events. p.. MK. After seven minutes of continuous screaming. All individuals descended rapidly from the tree and rushed towards the source. such as infanticidal attacks. Long-term data are collected on an interval basis. and their juveniles. NB’s adolescent daughter (NR) then took the carcass and initiated a play fight with an 37 . The females then moved into a vine tangle and. Results (1) Resident females attack an immigrant mother and kill her infant son (see also Reynolds 2005. approximately 1 week old.

Although it was too late to witness the actual killing. No attempt was made by any of the females to eat the infant. In this instance. The body was intact. The only adult male present during the whole event was ZF. The adult males ZF and MA were resting within 10m of NB and the carcass and never showed any interest in the carcass and never approached. and the adolescent male ran screaming to the presumed mother of the infant. allowing us to inspect it. vying for access to the body (NewtonFisher 1999b). NB. NB then regained possession and sat with the carcass. ML. each trying to drag and pull it off each other.Chapter 3: Lethal female competition adolescent male (KT). WL and KU fought over the body and may have been involved in delivering the fatal bite to the head. we consider it highly likely that FL was the infant’s mother who sustained her wounds whilst unsuccessfully trying to defend her infant son. KL. BN.42 NB moved off without the carcass. neither male showed any interest in the dead infant and 38 . ZN. and WL to inspect it but threatened FL and KZ when they tried to approach it. with puncture wounds to the head. the adult female (FL). who was lying in thick undergrowth with a fresh bleeding gash 2-3 inches long on her left upper arm. MK initiated it. She allowed KU. MA arrived after. When KT approached she threatened him. She is missing her right hand. we strongly suspect that the five adult females ZM. FL was a recent immigrant who came into the Sonso community about 10 months earlier with two older male offspring (approximately 4 and 9 years of age). NB retained possession of the carcass for just under an hour. making defence of an infant difficult. the perpetrators paid great attention to the dead infant. At 14. grooming it and keeping the flies away from it. Following male-led infanticides. Although we cannot completely rule out that the infant belonged to an unknown stranger female who disappeared during the incident. MK.

the young adult male BB approached the now westward moving screaming party.14 an old adult male 39 . KU. ZF descended and moved directly towards a party of screaming individuals. Then. pilo-erect. ZF moved away from the screaming party. When hearing screams about 30m away. who crouched on the ground screaming for several seconds. NK and ZF were following the females. chased by six resident adult females (NB.09 the stranger female managed to escape and ran approximately 25m away. the adult males BB. travelling towards another nearby feeding party. again piloerect. KW. we were confident that the males had not played any part in the lethal attack. produced SOS screams (Goodall. hitting at least one of the females. however the females quickly regrouped to continue the attack. they did not approach. WL and ZM). displaying. when we saw contact aggression between various screaming females.05. whilst the stranger female was lying on the ground shielding her infant from the ensuing attack. Approximately 3 minutes later. One male charged right through the females. At 13. pounded her back. The mother had a wounded and bleeding ano-genital region that was not swollen. RH. direct observations were not possible until 13.Chapter 3: Lethal female competition despite resting in the very close vicinity. thumping and stamping her. Katie Slocombe and field assistant Monday Gideon) were with the adult male ZF who was feeding in a Cynometra alexandrii tree. During this attack. we observed a stranger female with an approximately one-week old female infant. All but WL had infants under three years clinging ventrally. The females soon caught the stranger.00 hours. Due to poor visibility. 1986) about 5m from the females. Within the group of fighting females. buttress drumming and screaming continuously. (2) Resident females attack a stranger female and kill her infant daughter On 3rd February 2006. At 13. From these observations. pilo-erect. The adult male (NK) displayed close to the females and the young adult male BB. we (myself. at approximately 13.

now riding dorsally. At 13. They rushed towards the screaming party and found fresh drops of blood and part of an infant’s hand along the trail. with her male infant ZK.Chapter 3: Lethal female competition (MA) appeared and tried to disrupt the assault on the stranger by repeatedly forcing the females apart. Although in a state of partial decomposition and with significant injuries to the lateral side of the head and neck.00 we discovered the body of the killed infant in the adjacent tree. but the adult males NK. lifted the infant up.30. the chimpanzees returned to this same tree again and at 10. Two days later. She immediately chased BB.16. probably killing it. whilst ZF continued to display close by. and delivered a single bite spanning the head and neck. ZM sat on a log. At 13. The adult male MA was no longer visible.18 ZM. carried the limp infant by the pelvic girdle. The adult male MA then chased ZM away with the dead infant. The adult male BB immediately charged at her and hit her in the face with considerable force. two of our field assistants heard loud screams from the centre of the Sonso home range. the cadaver remained intact with no signs of attempted consumption.00. giving aggressor screams. the females successfully removed the infant from the mother and the resident alpha female (NB) ran along the ground carrying it. ZM was knocked backwards and hit the infant hard against the log as she recovered her balance. They also saw a low ranking 40 . but could hear them calling from a fruiting Ficus mucuso tree at 17. For the rest of the day. at approximately 09. The females continued to scream and compete for possession of the infant. ZF and BB continued to display. after considerable movement throughout their home range. 20m off the ground in a fork. (3) Resident females kill an infant On 2nd July 2006. The mother was no longer seen after this point. we were unable to follow the females as they moved through impenetrable forest.

who were involved in the previous killings. The occurrence of infanticides coincides with local peaks in the community sex ratio. as part of a large feeding party. using long term community composition data. meaning they may have been involved. less pronounced belly). Melissa Emery Thompson and field assistants could confirm that during the event. KW. In addition. all adult males of the community were approximately 800m away. screaming loudly. Further down the trail. known to be near her pregnancy term. RH.Chapter 3: Lethal female competition resident female (MK). Since the event occurred in the centre of the community’s home range and close to the research camp. Their whereabouts during the killing are unknown. it is very unlikely that males from the unhabituated neighbouring community were responsible for this case of infanticide or that an individual carried the infant from the periphery. retreating chimpanzees. suggesting that the infant was killed by resident females. was perpetrated by Sonso females. The rest of the infant’s body was never seen or recovered. KU. and ZM. The sex ratio of the community is also plotted over the same time scale.1 shows the timescale of the reported female-led infanticides in the last decade. were absent from this large feeding party. with no close relationships with other females and no offspring. a portion of maxilla with two unerupted incisors was found. The high ranking females NB. who is believed to move between Sonso and neighbouring communities. KL. as she was seen later in the same day with signs of recent parturition (sagging. (4) Timescale of infanticides Figure 3. like the previous incidents. We have strong reasons to believe that this. towards the sounds of the screaming. MK is a low ranking female. 41 . gaping sexual skin. We suspect MK may be the victim’s mother.

together. In all three cases. or strongly inferred. Relationship between the adult female and male sex ratio and the occurrence of female led infanticide in the Sonso community of wild chimpanzees. to be led by the resident females of the community and in at least two cases the same high-ranking females were responsible whilst the victims’ mothers were either a stranger or a recently immigrated female. In February 2008. Thus. ML. Arrows denote the occurrence of a female-led infanticide. the adult males were not involved in the killing. we encountered three such events in only 28 months. over the last decade. Discussion I have provided observational evidence for coalitionary lethal aggression in female chimpanzees. I conclude that these were purposeful. the remains of the attacked infants showed significant bites to the head.1. RH) were seen to attack 42 . Since July 2006. almost certainly the fatal wound in the first two cases. The three instances of infanticide described here were observed.Chapter 3: Lethal female competition Figure 3. In all cases. one further “attempted” infanticide has also been observed. high-ranking resident females (NB. It is also notable that. killings. not accidental. fulfilling a possible peace-keeping role. Instead in two cases they displayed in the vicinity or they even intervened and attacked the females to impede the infanticide.

why did the same high-ranking females participate. a well-known feature of chimpanzee behaviour (Clark-Arcadi and Wrangham. in the SE periphery of the Sonso home range (Monday Gideon. ruling out the idea of short-term nutritional gain as an explanation for this behaviour. ruling out intercommunity aggression as a feasible explanation. our females made no attempt to consume the infants after killing them. whilst this is relevant. 43 . Inter-community aggression: The females’ behaviour could be hypothesized as an extension of inter-community aggression. However. Pers. whilst injury to the mother seemed incidental. aggression to the mother terminated as soon as the infant was taken. Moreover. it is not included with the three previous cases as the survival of the offspring is unknown. However. such as the well-known Gombe cases (Goodall 1986).Chapter 3: Lethal female competition and attempt to remove the ~1week old infant of an immigrant female. What was the impetus behind these infant killings and attempted killings. the most recent victim was a low ranking resident of Sonso and in the observed cases.comm). and why did some of the males directly or indirectly attempt to disrupt the killing? I can think of several explanations. 1999). our females were clearly fixated on removing the infant from the mother in order to kill it. some of which can be ruled out by our observations: Nutritional gain: In contrast to the other instances of female-led infanticide.

presumably because they increase their mating success.1). Contrary to traditional views of male chimpanzee behaviour. the Sonso community has had an unusually high influx of female immigrants who have brought one or more dependent offspring with them. It is also relevant that the community is probably unable to expand its current home range to access additional resources because the number of adult males has simultaneously gone down. 2005). several studies have found that the home range of the Sonso community has a particularly rich and stable food supply with relatively little seasonal variation (Newton-Fisher. An estimated 5-10 parous females immigrated between 2001-03. Concerning food. Second. like other communities the Sonso chimpanzees have a female-skewed sex ratio.Chapter 3: Lethal female competition Intra-group female competition: Increased competition is a more likely explanation for these attacks. Males are not normally hostile to these immigrants. 2003. 2003). heightening the chance that 44 . but this has now increased from approximately 1:1 in 1996 to almost 1:3 in 2006 (figure 3. with a total of 13 stranger females having now taken up residence (Emery Thompson et al. 2006). However. In what ways could heightened female density provide a competitive challenge for resident females? Access to food and sexual partners are promising candidates. so competition for access to highquality core feeding areas may still be considerable. In the last few years. perhaps making this community particularly attractive for immigrants.. The recent influx has dramatically increased the community size in a short period of time. but it is difficult to assess the relative importance of each. Bates. including males as old as 10. the home range of this community is relatively small (Newton-Fisher. however. they have not shown any serious aggression towards the offspring of these females. suggesting that a critical threshold may have been surpassed. and this number increased even further from 2004-2006.

. These females are generally more competitively able than lower-ranking females (Pusey et al. 1985) have been observed to intervene during female-female aggressive interactions.. the adult male MA. possibly compromising resident females in their ability to mate with preferred males (Goodall. in press) and should thus be most interested in maintaining their access to such resources. 2007. suggesting that he may have copulated with her during that time. The two peace-keeping males. 1996). who showed the strongest motivation to stop the March 2006 infanticide. adult males actively attempted to disrupt. not advance. In contrast. It is also relevant that most females that participated in these killings were relatively old and high-ranking individuals (Newton-Fisher. We never observed any participation of the adult males in these infanticides. 1986).. were joined by higher- 45 . 2008a). It is possible that this was a direct result of males recognising the stranger female as a previous or potential mating partner. the progression of the infanticide. however such mediation behaviour has not yet been documented in wild chimpanzees. 1987) and gorillas (Hoff. compounded by the fact that the three observed infanticides occurred when the adult female:male sex ratios were highest suggests that competition for male chimpanzees could be a real challenge for females in this community. 2005). 2002. MA and BB. when the sex-ratio is skewed.. Indeed. Pfefferle et al. males and their sperm can become a limiting resource (Brauch et al. The observations presented here suggest escalated aggression may be one such strategy. This. Wedell et al.Chapter 3: Lethal female competition several females will be in oestrus at a given time. was not sighted with the rest of the community during the stranger female’s time of conception (July-August 2005). Captive male orangutans (Zuker. 1997. Male peacekeepers The behaviour of the adult males during these infanticidal attacks is also of interest (see Wrangham & Peterson. Recent research has indicated that in promiscuous mating systems.

NK and ZF. recent observations of very similar behaviours have now been documented from other study sites (Gombe: Pusey et al. in press a. 2008). It is therefore possible that these males have never had contact with the stranger female. past copulatory experience appears to be the best explanation for the unusual peace-keeping behaviour in some males. this aggression. 46 . Coalitionary and severe female aggression may not be as rare as previously thought. in press. these males would enjoy no immediate fitness benefits from intervention although there are numerous other explanations for why they remained neutral during the conflict. the emerging picture is that the Sonso females respond with extreme physical violence to increased intragroup female-female competition caused by high rates of immigration.. but these males only displayed in the vicinity of the event without directly intervening.Chapter 3: Lethal female competition ranking males. Mitumba: Pusey et al..b. in press. which is probably part of the wider female behavioural repertoire.. in fact. in press. was facilitated by an extremely skewed sex-ratio and high density. These observations provide further evidence that chimpanzee females are able to form alliances with non-kin (Emery Thompson & Wrangham. 2005). Thus. This may have led to increased food and reproductive competition between females. Concerning the females’ behaviour. Newton-Fisher. To summarise. For female chimpanzees such competition may therefore represent a significant threat associated with social living. Kanyawara: Kahlenberg et al. Tai: Boesch et al. In the case of Budongo however.. in this instance to attack other females and their infants.

Hormone analysis further showed that the calling behaviour of copulating females was unrelated to their fertile period and likelihood of conception.Chapter 4: Copulation call production Chapter Four: Female chimpanzees use copulation calls flexibly to prevent social competition Summary The adaptive function of copulation calls in female primates has been debated for years. consisting of relatives. In chapter three I have shown that competition between females can be dangerously high in wild chimpanzees. females called significantly more whilst with highranking males. I analysed the copulation calling behaviour of wild female chimpanzees at Budongo Forest. The fact that the female is usually part of a complicated social network.. is not normally considered. which enables females to advertise their receptive state to males. but suppressed their calls if high-ranking females were nearby. Zuberbuhler K. In this study. confuse paternity and secure future support from these socially important individuals. Instead.. but found no support for the male-male competition hypothesis. One influential idea is that copulation calls are a sexually selected trait. Female chimpanzees use copulation calls flexibly to prevent social competition.W. Female primates are now known to play crucial roles in the social fabric of the group. and my results here indicate that females use their copulation calls strategically to minimise the risks associated with such competition. The results from this chapter have been published in the following paper: Townsend S. and it is therefore somewhat surprising that this has never been properly considered as a selection factor in this reproductively important vocalisation. Male-male competition ensues and females benefit by getting better mating partners and higher quality offspring. PLoS ONE 3(6): e2431 47 . Deschner T. (2008). offspring and social competitors. Uganda. whilst these adaptive explanations address a number of important theoretical points their focus is on how the female’s calls affect the nearby male mating partners. However. Copulation calling may therefore be one potential strategy employed by female chimpanzees to advertise receptivity to highranked males.

Semple. copulation calls are loud. acoustically distinctive vocalisations emitted prior to. Indeed. 1996. considerable debate has surrounded the adaptive significance of these conspicuous acoustic signals. although it is unlikely that any one hypothesis in isolation is sufficient to explain call evolution. to the receptive condition of the female caller (Cox and Le Boeuf. Semple. however in Old World monkeys and apes.g. The most common hypothesis invoked to account for the evolution of such calls is that they are sexually selected traits to alert males. A number of different hypotheses have been put forward to explain the adaptive significance of copulation calls (Pradhan et al. Pradhan et al. 2001. 1988].Chapter 4: Copulation call production Introduction In various animal species copulations are accompanied by a distinct vocal behaviour. it is more commonly females that vocalise (Hauser. 1998b). 1996. other than the mating partner. 1972]. Hauser. Due to their prevalence. African elephants (Loxodonta africana) [Poole et al. the female ends up with the most dominant partner (Henzi. 2006). which are not mutually exclusive (Semple et al. 1995. 1977) can operate at two distinct levels. Firstly. Interestingly. Calls can be produced by both males and females participating in the copulation. not all copulations are accompanied by calling behaviour. Copulations accompanied by a 48 ... 1994. Oda and Masataka. 1977].. suggesting that females have some control over call production. O’Connell and Cowlishaw. 2002). In primates. 2006). indirectly.. calls may operate to stimulate overt competitive interactions between males so that. 2001) with the result of inciting competition amongst them. 1990. 1996). The “incitation of male-male competition” hypothesis (Cox and Le Boeuf. 1998b. the copulation call (e. during or just after copulation. lions (Panthera leo) [Schaller. 1977. and humans (Homo sapiens) [Hamilton and Arrowood. copulation calls may operate at more than one level with multiple functions (Semple. elephant seals (Mirounga angustirostris) [Cox and Le Boeuf. 1978].

females should call to advertise ejaculation (O’Connell and Cowlishaw. but they will be particularly motivated to mate with the female shortly after a successful mating by another male. and this could generate additional benefits for the female due to sperm competition (O’Connell and Cowlishaw. suggesting that there are strong selective pressures acting on females to evolve behavioural or sexual counter-strategies to protect their infants: copulation calls may well be one such counter-strategy. 1994) and such calling should lead to a decrease in the time period between successive matings (Semple. 1994). Secondly. 1998b). less desirable males and increase subsequent levels of male aggression.. males do not attempt to prevent insemination per se. the desired empirical support is weak. Instead. 1990). however. females are primarily interested in receiving copulations from as many socially important partners as possible. especially for 49 . Although the theoretical reasoning behind the incitement of male-male competition and the paternity-confusion hypotheses is sound. Polyandrous mating. In many primate species females are notoriously vulnerable to infanticide (Van Schaik. If female calling behaviour has been shaped by sperm competition. the paternity confusion hypothesis makes no predictions about females trying to increase the quality of partners or sperm. 1994).Chapter 4: Copulation call production call are then predicted to primarily occur with low-ranking. safeguarding them from their infanticidal tendencies and gaining their future support. copulation calls may lead to multiple mating partners. and sperm competition that follows from it. In contrast to the male-male competition hypothesis. 2000. 2007). and it has been argued that this lowers the risk of male infanticide (O’Connell and Cowlishaw. Under this scenario. Aggressive interactions can also occur during or after copulation to prevent insemination or future matings (Hauser. increases paternity confusion for individual males. Muller et al.

. Henzi. or confuse paternity amongst multiple males.. Nikitopoulos et al. is far more pervasive in chimpanzee societies than previously thought (See chapter three. Pusey et al. Very little is known about the degree to which female primates adjust calling behaviour in relation to the composition of the audience.. which are typically matrilineally bonded (Hauser. in contrast to male-bonded chimpanzees. 2008a). Most empirical work so far has been done with different monkey species. 2002). For example female-alpine accentors. less competitively able females. in press a. it may thus not be beneficial to advertise successful matings with copulation calls if other females are nearby.. Female competition may therefore represent a significant social pressure in the lives of wild female chimpanzees. it has also been suggested that copulation calls change based on female sexual status (O’Connell and Cowlishaw. 1998b. males and their sperm may be one such limiting resource (Pfefferle et al. sing only during their fertile timeperiod (Langmore et al. Females are likely to compete with each other over access to resources and in mating systems where promiscuity is high. and (b) whether the desired mating partners are present in the audience. but hormonal data are not usually available to determine the precise time of ovulation. 1996. Prunella collaris. 2007. b. O’Connell and Cowlishaw. In primates. Semple et al. For lower ranking. A number of studies have investigated the influence of the female reproductive stage on vocal production. 1997. 1994. 1992). in press).Chapter 4: Copulation call production chimpanzees. A second relevant point is that if copulation calls function to increase a female’s reproductive success. Semple. 2004). then it is reasonable to predict that callers should take into account (a) at which stage in their cycle they are.. 1994. and the aggression that accompanies it. Kahlenberg et al. especially if this increases the likelihood of 50 . 1993. A growing body of evidence suggests that femalefemale competition. 1996) and the stereotyped 50kHz vocalisations produced by female brown rats are only given during pro-oestrus (Matochik et al...

the group comprised 78 individuals including 8 adult males and 25 adult females. behaviour and determination of female swelling size Around the time of ovulation.Chapter 4: Copulation call production aggression. Of the 25 adult females. female chimpanzees exhibit sexual swellings. 1990). Materials and Methods Study site and animals I studied the sexual behaviour of a number of individuals belonging to the Sonso community of the Budongo Forest. Goodall. 1979. and to minimise the effects of social competition caused by other females on the other hand. Copulation calls. Uganda. at the time of this study. I conducted a study on the copulation calling behaviour of wild female chimpanzees from the Sonso community of the Budongo Forest. Hasegawa and Hiraiwa-Hasegawa. Based on these considerations. As mentioned in chapter two.. To address these points. Our pilot observations revealed that females often remained silent during copulations. 1979. I hypothesised that females adjusted their copulation calls. Copulation calls consist of a rhythmic succession of high-frequency 51 . 1990). 1986). during two field seasons (January 2006April 2006 and October 2006-March 2007). data were collected from 7 adult females. Deschner et al. Female chimpanzees mate promiscuously with multiple males (Tutin. 1983. Three additional females also had sexual swelling and copulated during the study but were excluded from analyses due to low copulation frequency (<15 copulations). although the reasons for this behaviour remained largely unknown (Hauser. 1986). to maximise paternity confusion by soliciting copulations from nearby males on the one hand. 2003) and females` sexual activity is usually confined to this period (Goodall. The average duration of the maximum swelling period is about ten days (Tutin. Uganda (Reynolds 2005). but they do not produce copulation calls every time (Hauser.

Copulation calls have a frequency range of 700-1000Hz. after mounting and intromission (figure 4. Maximum tumescence was determined following Furuichi’s (1987) method.1). rather than labial occlusion (Dahl. Copulations from cycling adult females were collected using all day focal follows on each day of the female’s maximum tumescence phase.5s and (B) a single copulation call of approximately 0. Given that only one female could be followed for this duration. Only copulations occurring during the maximum tumescence phase were considered. ad-libitum observations of copulations were also taken. 1999). (C) The lowest visible band is the fundamental frequency from which acoustic measurements were taken with three visible harmonic bands. Frequency resolution: 86. Depicted is (A) a complete copulation calling bout of approximately 6.6s by the female JL. Figure 4. yet more than one female could cycle at any one time. Filter bandwidth: 159Hz.1Hz. Copulation calls can be reliably identified by human observers and are audible in forest habitats up to about 50m. In addition to swelling 52 .Chapter 4: Copulation call production squeaks or screams and typically begin during the copulation. Inter-observer agreement between myself and field assistant Monday Gideon (MG) was a pre-requisite for final assessment of female swelling size.1: Time-frequency spectrogram of a female copulation call from Budongo Forest during maximum tumescence. which uses degree of wrinkling of the sexual swelling (on a 4 point scale at Budongo) as the main parameter. Sexual skin swelling characteristics were recorded every morning through visual inspection of the perineal area.

. respectively. samples were analysed by Tobias Deschner and colleagues (MPI-EVAN). calculated from replicate determinations of quality controls were. I collected regular urine samples during the period of maximum tumescence. Serial dilutions of urine samples of the follicular and luteal phase gave displacement curves parallel to those obtained with the appropriate standard. figure 4.. Only calls that occurred during the copulation were considered to control for the vocalisation being elicited by an alternative stimulus other than the copulation. all hormone levels were divided by the urinary creatinine concentration as described in Bahr et al. and assessment of the fertile period To determine approximate timing of ovulation and its potential effects on calling behaviour.26% (low). hormone analysis. duration to next copulation and composition of the audience during copulation. Intra.and inter-assay coefficients of variation. presence/absence of copulation call. unfortunately there was not enough time to conduct assays alone.Chapter 4: Copulation call production size. Urine sample collection. aggressive behaviours following a copulation.52% (high) and 13. temporal occurrence of call in relation to copulation. using enzyme immunoassay procedures (Heistermann.94 and 6. To compensate for variations in the volume and concentration of urine samples. 2003.31 and 11. (2000). Although I did familiarise myself with the laboratory procedure.5 pg. for immunoreactive pregnanediol glucuronide (PdG). Based on the defined postovulatory rise in PdG levels. 1996). 2004). 7. with sampling gaps of no greater than two days.2). The sensitivity of the assay at 90% binding was 12. the day of ovulation was presumed as the day preceding the day of PdG increase (Deschner et al. Therefore. They were stored in 2-ml polypropylene Cryotubes in liquid nitrogen until shipment on dry ice to the laboratory. Based on human data regarding the survival time of 53 . Samples were collected directly after an individual had been observed urinating by aspiration of the urine from plastic sheets or vegetation using disposable plastic pipettes. I noted the following variables: identity of mating partners.

with the post-ovulation period being the period of maximum tumescence following POP.2: Profiles of urinary pregnandiol in ng/mg creatinine and perineal swelling for the adult female WL during April 2007. provided I could identify a target of aggression. defined as the day of ovulation plus the three preceding days. such as arm raises or displays. 2003). PrePOP: period of maximum tumescence prior to POP PostPOP: period of maximum tumescence following POP. and during the subsequent 10-minute time-window. Aggressive events could range from ‘mild’. Figure 4. to ‘severe’. such as chases or stamping and beatings (Slocombe and Zuberbuhler.. the fertile period (POP) was defined as the day of ovulation plus the three preceding days (Deschner et al. Behavioural observations Male-male aggression I scored all instances of aggression during copulations.Chapter 4: Copulation call production ovum and sperm. POP: periovulatory or fertile period. 2007). 54 .

5 high-ranking males (NK.Chapter 4: Copulation call production Sperm competition In wild chimpanzees. BB. Boesch and Boesch-Achermann. The prediction is that sperm competition increases as the time interval decreases. ZF. dominance rank is usually assessed by using the occurrence and direction of pant-grunt vocalisations. it made little sense to attempt to construct a linear dominance hierarchy.. 1979. De Waal. 1998b). MA) and 3 low ranking males (GS. Beilharz and Mylrea. The direction of these vocalisations is regarded as a good indicator of relative social status (Bygott. Eight of the community adult males received pantgrunts from other males. 2006). allowing us to assign each individual with a ‘dominance value’ (DV=arcsine of the square root of the proportion. Because of the instability of the male hierarchy at the time of study. 55 . 1963). Juvenile and sub-adult males were not observed to receive any pant-grunts. Sperm competition has alternatively been assessed indirectly. NewtonFisher. 2006). 2000). it is difficult to determine reliably whether or not ejaculation has occurred (O’Hara and Lee. BO). There were two clusters of individuals (see table 4. Male rank Male rank is known to affect the production of copulation calls in many primate species (Pradhan et al. DN. MS. In chimpanzees. Instead I determined the status to each male by calculating the proportion of other males in the community from whom he received pant grunt vocalisations.1) with similar DV scores. 1982. by measuring the time interval between successive copulation events (Semple. 1997.

2 0.48 0. Rank relations between female chimpanzees are more stable than 56 . my field assistant and I conducted an additional and more detailed search of the area to account for individuals that might have joined or left the party since the previous scan. Female rank Female rank was determined in a previous study of female-male aggression (Newton-Fisher.14 High High High High High Low Low Low Audience effects Previous work has shown that wild chimpanzees can adjust call production depending on who is likely to listen to their calls (Slocombe and Zuberbuhler. see table 4.14 0. consisting of a mother and her dependent offspring. usually around 10 individuals. 2007). This was particularly important for copulations that occurred in trees.38 0. with individuals joining or leaving regularly. Party composition is relatively fluid. and group members are often not in direct visual contact. Male NK DN ZF BB MA MS BO GS DV Score Rank group 0. 2006. or in mixed-sex parties of different sizes. A party was defined as any individual within a 50m radius (Reynolds. To determine whether the audience had an impact on copulation calling I noted party composition at 15-minute intervals when following a female.48 0.1: Adult males.2).45 0.5 0. Every time a copulation event occurred. They usually travel in small family groups. 2005) of the focal female.Chapter 4: Copulation call production Table 4. dominance value scores and rank groups for the adult males of the Sonso community over the study duration. where the female has a better observational vantage point than observers on the ground.

Z = -2. before and after transformation.791. 2007). whether they are normally distributed.05. Logistic regression makes no assumption about the distribution of the independent variables.016. I used non-parametric statistics.. male rank and male audience number (Wilson et al.1 (R Core Development Team. 15. female audience composition. For small sample sizes. The females were more likely to produce copulation calls when they mated with high-ranking adult males than low-ranking males (Wilcoxon exact test N females = 7. p=0.2).37. Because the calling behaviour data were scored as a binary response (ie “yes” or “no”) and I was interested in the independent and potentially interactive effect of the different explanatory variables. Results Mating behaviour of female chimpanzees All seven monitored females gave copulation calls during mating. If the data failed to meet conditions for parametric analyses. I calculated exact p-values. 2001). that is. 2003). figure 4. All tests were two-tailed and significance levels were set at α= 0. Statistical analyses Whenever possible I conducted parametric analyses.Chapter 4: Copulation call production between males (Wittig and Boesch.5. I also employed a binary logistic regression. figure 4.3). The following variables were incorporated into the model: female ID.4).0 and R version 2. All described statistical analyses were done using SPSS v. There was no difference in 57 . as recommended by Mundry and Fischer (1998). and there was no evidence of any significant changes since that study. but only in a minority of cases: The females copulated a total of 287 times and produced copulation calls during only 104 (36%) of copulations (table 4. with all seven females showing the same pattern (Cronbach’s alpha test for reliability = 0. linearly related or of equal variance within each group.

Chapter 4: Copulation call production calling behaviour when females copulated with low-ranked adult males and even lowerranking subadult males (Wilcoxon exact test N females = 7. Z = -0. 2005). Uganda (adapted from Newton-Fisher. Table 4. Bar graphs showing the percentage of copulations accompanied by calls given by seven females when copulating with high (N=5) and low (N=3) ranking males. 58 . p=0.405.2: Rank and copulation calling behaviour of seven adult females of the Sonso community. Budongo Forest.3: Copulation calls and the effect of male rank.813). Female Female rank Total number of copulations % copulations accompanied by a call LL WL NB MK KU KY JL Total Low Low Alpha Low Low High Mid 66 68 50 37 18 29 19 287 53 34 38 35 11 24 26 36 Figure 4.

Chapter 4: Copulation call production Figure 4. twotailed). p=0. p=0. but remained silent. Out of the nine instances of aggression following a vocalisation. two-tailed. These four cases of female-caused aggression were particularly severe and always directed at a lowranking female. If the same four low-ranking females copulated. 35 (12%) observed copulations elicited aggression by a third party individual. There was no difference in the occurrence of aggression after silent or vocal matings (controlled for copulation number: N silent=26. then the high-ranking females never responded with aggression. three by high-ranking males. and two by low ranking males.2. either leading to interruption of the copulation or to targeted aggression to one of the mating partners within 10 min. Nsilent = 4. binomial test (0.02). Nvocal =4.4: Individual variation in copulation calling behaviour.63). In no case were they likely to see the copulation event (Fishers exact test. Line graphs showing the proportion of copulations accompanied by a call when copulating with high (N=5) and low (N=3) ranking males for each of the seven females. 59 . four were caused by high-ranking females. N vocal=9.

N=117. Z=-0. during the fertile peri-ovulatory period (POP) and after ovulation (Post-POP). I found no relation between the time interval between successive copulations with different males and the likelihood of a female producing copulation calls (Wilcoxon exact test.45. Mann Whitney U test: U= 536. Mean femalesvocal = 2. N=126 total. females called less the more adult females were in the 60 .844).3. figure 4. NB: N=2). Nfemales = 7. Because one female (WL) did not exhibit a Post-POP period and another (NB) did not exhibit a Pre-POP period.10. Wilcoxon Exact test Z=-1.Chapter 4: Copulation call production Finally. p=0. i. Females called prior to the fertile peri-ovulatory period (Pre-POP). which allowed us to determine the exact time of ovulation. WL: N=3. In contrast.344. p=0. only five cycles were included for each analysis. p=0.5). figure 4. N=121. Nsilent =6. p=0. we were able to analyse the hormonal profiles of six complete oestrus cycles (LL: N=1. I randomly selected for each of the seven females an equal number of copulations (N=18).328.314.5). p <0. Nvocal = 28. I found no significant difference in the calling rate between Pre-POP and POP periods (Binomial GLMM with female ID as a random factor Z=-0. Adult male audience size had no effect on call production by the copulating female (Mean malessilent = 3. which were subjected to analyses. Z=-1. the number of adult females in the party had a significant effect on call production (Mean femalessilent = 3.41.916. Nfemales= 5.789.6. Nsilent = 62. t=-4. Audience effects To test for audience effects. p=0.181).04. Hormonal analyses In collaboration with Tobias Deschner. Nvocal =6. Nfemales =7. Mean malesvocal = 3.e. despite the fact that there were consistently more high-ranking males present when a female copulated with a high-ranked male (Paired T test.430) or between POP and Post-POP (Binomial GLMM with female ID as a random factor.001).

Nsilent = 62. I was particularly interested in how female audience composition affected calling behaviour. Of the variables tested male rank. Nvocal= 28. Mean femalesvocal = 2. Both were high-ranking females and there were indications that they behaved differently in the presence of other females. Male audience (Nfemales=7. who were excluded from this analysis. The observed audience effect was mainly driven by the social position of listening females. Ncopulations=126): the number of adult males present. Nfemales = 5.Chapter 4: Copulation call production party.01. Mann Whitney U test: U= 516. Error bars represent Mean +-1 SE. KY). Sample sizes were too small for two females (NB. female audience 61 .5: Mean number of individuals in the audience in the presence/absence of a copulation call. p=0. compared to the other five lower-ranking females. Figure 4. Female audience (Nfemales = 5. Ncopulations=90): the number of adult females present. To identify the independent and potentially interactive influence of the determining variables I conducted a binary logistic regression.025).38. Females called significantly less if they were surrounded by a female audience that contained individuals of equal or higher rank than themselves (Mean femalessilent = 3.

male rank p= 0.6).7).421. Figure 4. Error bars represent Mean +-1SE. p= 0.004. The model explained variation in female calling behaviour with 82% accuracy. two-tailed). The significant interaction effect suggested that the females’ response to female audience composition also depended on the rank of the male mating partner. this was most apparent when females copulated with high-ranked males (figure 4.Chapter 4: Copulation call production composition and male rank*female audience composition explained a significant proportion of the overall variance (binary logistic regression with female ID as a random factor χ2 = 8.029. male rank*female audience composition p=0.6: Line graphs showing the mean number of high and equal ranked females in the audience when copulating with a) high ranked males and b) low ranked males. female audience composition p=0. Whilst there was a trend to call less when more high-ranking females were in the audience for both rank groups. 62 . N=90. a rate significantly higher then that when running the model with no explanatory variables (Binomial (0. Nagelkerke r2 =0.024.595.043).

Females call significantly more when copulating with high. 1977) despite its prominence in the sexual behaviour literature. Specifically.. and since other dominant males are usually nearby in these circumstances. then why do females call? My results suggest that other social variables are important in driving these vocalisations. 1995. 2004). but suppressed calls if high-ranked females were present. females did not produce calls when mating with low-ranked males in order to instigate disruption by high-ranked individuals (Nikitopoulos et al.compared to low-ranked partners. Nikitopoulos et al. 2002. they do not provide any information about the timing of their ovulation. copulation calls did not lead to increased levels of aggression towards the mating pair. they called more when mating with higher-ranked males. In this sample. Pradhan et al. Instead. Semple et al. this study lent no support to the ‘male-male competition’ hypothesis of copulation calling (Cox and Le Boeuf.. Although females appear to be motivated to advertise their receptivity. I also found no evidence that chimpanzee copulation calls operated at the ‘sperm competition’ level (O’Connell and Cowlishaw. 1994).Chapter 4: Copulation call production Discussion Overall. calling is one potential strategy allowing a female to signal her receptivity to a large audience of high-ranked males. Instead. 1994. 2004. 2006). an effect also reported in other primate species (O’Connell and Cowlishaw. females produced copulation calls preferably when mating with highranked adult males. Whilst it was virtually impossible to determine the occurrence of ejaculation. Hormonal analysis showed that female calling behaviour was unrelated to their fertile period and likelihood of conception. If the male-male competition hypothesis does not explain copulation calling behaviour. the duration to the next copulation was unrelated to the female’s calling behaviour. Oda and Masataka.. a pattern that also holds for Barbary macaques where precise information on the timing of 63 ..

64 .. 2008a). apart from trying to increase paternity confusion. My data suggest that lower ranking females refrained from calling when mating with high-ranking males if high-ranking females were nearby.. In this context. By calling in the presence of high-ranking adult males and by concealing ovulation. Mahale (Nishida. Research from long-term field studies increasingly shows that chimpanzee females are exposed to severe social pressure from other group members. Possibly because of their previous mating history. particularly amongst socially important males. females may prevent monopolisation by a single male and to increase paternity confusion amongst other males. it reduces the probability that males will attack infants potentially sired by them (O’Connell and Cowlishaw. has a two-fold advantage. Firstly. high-ranking males have been observed to intervene during female aggressive events. My data are consistent with the idea that chimpanzee females use copulation calls as a vocal counter-strategy to minimise these threats.Chapter 4: Copulation call production ovulation is not available in copulation calls (Pfefferle et al. in press). at Gombe (Pusey. Pusey et al. suggesting that they were trying to conceal their sexual activity in these circumstances. confusing paternity. 2006). Secondly. 1980. including during femaleinitiated agonistic encounters. 1989) and Budongo (see chapter three). In chapter three I have shown that female chimpanzees can suffer substantially from infanticide-related threats. 1994). it is likely to improve a male’s general willingness to provide support. Chimpanzees produce copulation calls at much lower rate than other primates (Pradhan et al. especially when resources are limited.. suggesting females take other factors into account. which in some cases have resulted in female-led infanticidal attacks.

Chapter 4: Copulation call production Unlike most other primates, chimpanzee females leave their natal group at adolescence to immigrate into neighbouring communities. Immigration can affect the adult sex ratio of a group (see chapter three), increasing competition for resources between females, such as highquality foraging areas (Pusey et al., 1997; Kahlenberg et al., in press a), and possibly the amount and quality of available sperm (Nakatsura and Kramer, 1982; Dewsbury, 1982; Van Voorhies, 1992; Gage and Cook, 1994; Preston et al., 2001; Weddell et al., 2002; Marson et al., 1989). As a consequence, more competitively able high-ranking females should have an interest in maximising their own access to such resources and escalated aggression may be one strategy (see chapter three, Pusey et al., 1997; in press; Kahlenberg et al., in press b). One counter-strategy for lower-ranked females is to form short-term associations with the adult males of the community (Kahlenberg, Pers. comm.; in press b) and, as suggested by this study, to modify their copulation calling behaviour (De Waal, 1982) when high-ranking resident females are likely to witness their sexual activities.

To conclude, female copulation calls in primates and other groups of animals have usually been interpreted as male-directed signals, for example to advertise fertility and incite malemale competition, but my findings in wild chimpanzees do not support this view. In this study, chimpanzee females adjusted their calling behaviour in flexible ways, potentially to avoid aggression from other females and possibly to secure future benefits from the socially important males. Data from more females and different study sites will be required to test this hypothesis more thoroughly (see chapter seven). For many years, female chimpanzees have been regarded as the more peaceful sex. However, there is increasing evidence from a number of communities studied in the wild, which indicates that female competition plays an important role in dictating female behaviour and these data provide further support for this view. My results indicate that the social pressures deriving from resource competition have


Chapter 4: Copulation call production acted as an important selective force, shaping the copulation calling behaviour in wild chimpanzees.


Chapter 5: Copulation call structure Chapter Five: Copulation calls in female chimpanzees convey social information

Copulation calls are a relatively common feature of female primate behaviour. As shown in the previous chapter, wild oestrous chimpanzees produce these calls probably to avoid monopolisation by single males and to minimize aggression by other females. In this chapter, I investigated to what further degree such social demands were reflected in the calls’ use and acoustic structure. Although females were generally reluctant to produce copulation calls when mating with low-ranking males, they called regularly in trees where furtive mating with these males was not possible. Hormonal analyses showed that the acoustic structure of copulation calls did not covary with the ovulatory status of the calling female, further confirming my previous findings that the calls functioned to confuse paternity. However, the calls’ acoustic structure reliably encoded the identity of the calling female and to some extent whether her male partner was of high or low social rank. In sum, chimpanzee females may be providing honest information to other group members about their sexual activity although they appeared to act strategically as to when to provide this information.

The results from this chapter have been submitted for publication: Townsend, S.W, Deschner, T., & Zuberbuhler, K. (Submitted). Copulation calls in females chimpanzees convey social information.


chimpanzee offspring depend on their mothers for more than a decade and subsequently both rely on males in a number of contexts. regardless of their exact fertile phase. or during competition over resources (Kahlenberg et al. 1998b. However in the previous chapter I have shown that male-male competition is probably just one of a number of social factors that influence reproduction and its accompanying behaviours. Reynolds. 1994. Given that male rank relations can change rapidly with young males typically rising through the ranks in a few years. Since high-ranking males have the power to avert infanticidal attacks by other group members (primarily females at Budongo). 2000. raids from neighbouring communities (Mitani & Watts. such as encounters with predators (Boesch. 68 . possibly to confuse paternity and enlist future support from socially important males against aggressively competitive female chimpanzees. choosing to mate with young lowranking males may be part of a successful longer-term social strategy for any female. Reynolds. Infant chimpanzees are most vulnerable in the first few months after birth. particularly due to infanticide by other group members (Goodall. confusing paternity more widely amongst the males (Goodall. Boesch and BoeschAchermann. see chapter three). 2005).. 1986. 1991b).. Females called most when with high-ranking males. 2002). 2007. 1986. Semple et al. 2005). inciting competition amongst them (O’Connell and Cowlishaw. 1996. oestrous females should thus be particularly interested in mating with them.Chapter 5: Copulation call structure Introduction In many primates copulation calls are thought to alert nearby males to the female’s receptive state.. in press a). 2005). However. Semple. especially if males use their mating history as a measure of paternity certainty (Murray et al. Henzi.

Hauser. 1998. a behaviour also observed at other research sites (Gombe: Goodall 1986. high-ranking males tend to aggressively exclude lower ranking males from mating. to date.Chapter 5: Copulation call structure In all chimpanzee communities. Encoding social information within the acoustic structure of these calls is one potential mechanism. natural selection is likely to favour mechanisms that enhance any such function. in order to copulate silently with a low-ranking male and then return to the rest of the group (Reynolds. This female strategy sets up an inevitable conflict of interest between males and oestrus females. Boesch and Boesch-Achermann. This is a vital piece of the puzzle 69 . Despite this potential. 2006. matings with lowranking males are ostensibly cryptic and females often refrain from producing copulation calls. animal vocalisations are capable of encoding varied information sets. Copulation calls play an important role in the lives of female chimpanzees helping them to avoid social conflict and secure reproductive success. females sometimes call during copulations with low-ranked males. 2004. usually by harassing the female (Goodall. Van Schaik et al. As described in chapter one. Watts. 2000). 1990. 2007). Perhaps due to this additional social threat. a finding that was hard to explain from the previous study and something that I was particularly interested in investigating. 2000.. 1986. females have been seen sneaking away from the main party. Given the unusually long chimpanzee inter-birth intervals in comparison to other non-human primates (Tutin. Kahlenberg.. As a result. oestrous females have been observed to mate with lowranking males. Duffy et al. 1979). Tai: Boesch and Boesch Achermann. most copulation call studies have focused primarily on the contexts eliciting production with little attempt made to assess what additional information may be encoded within the acoustic structure of these calls. 2005). At Sonso. However.

more broadly. females should behave strategically when mating with low-ranking males (Pradhan et al. Muller et al. as already described in female baboons (Semple.. I analysed 70 . chimpanzees typically travel in family units or small parties of less than a dozen individuals and individuals rarely have direct visual contact with all other group members. In this study. in addition to call rates. I thus predicted that natural selection should act against signalling the fertile period as this would lead to monopolisation by the most powerful male and decrease paternity certainty of others (Pradhan and van Schaik. In their natural forest habitat.. if this is not.. 1986. I expected females to provide acoustic cues that revealed their individual identity. 2002). 2006). how they function in their daily lives. Since females sometimes overlap in their oestrous cycles and because their relationships are individualised (Goodall.Chapter 5: Copulation call structure that needs to be addressed if we are to better understand exactly why female chimpanzees produce copulation calls and. produce copulation calls but honestly signal the rank of the male with whom they are mating (Semple et al.. when conception is impossible (Pfefferle et al. if high-ranking males aggressively exclude low-ranking males from mating. I was interested in elucidating what information copulation calls conveyed to the listening audience and how this related to the adaptive function of these vocal signals. One possible strategy for females therefore is to remain silent if furtive mating is possible or. 2008). 2006). 2008a). Finally. To systematically investigate the potential information content conveyed in these signals and to further understand in what ways female chimpanzees use their copulation calls. there should also be no acoustic differences in copulation calls given by a female during her ovulatory phase and after ovulation. In the previous study (see chapter four) it was shown that female chimpanzees tried to maximise paternity confusion amongst males. 2001). Hence.

hormone analysis. MA) and 3 low ranking males (GS. additional ad-libitum copulation call recordings were taken. During this period. For more details regarding hormonal analysis see chapter four. ZF. Male rank As explained in chapter four. BO). BB. Methods Study site and animals I studied a number of individuals residing in the Sonso community of the Budongo Forest. Tobias Deschner and colleagues could reliably identify the day of ovulation and hence the female’s fertile period. I collected regular urine samples during the period of maximum tumescence. Juvenile and sub-adult males were not observed to receive any pant-grunts. Through Enzyme immuno-assaying (EIA) PdG levels in urine. 5 high-ranking males (NK. There were two clusters of individuals with similar dominance value scores (see chapter four). DN. Urine sample collection. 71 . Recordings of calls were collected from 6 adult cycling females using all day focal follows throughout the female’s maximum tumescence phase.Chapter 5: Copulation call structure the acoustic structure of wild female chimpanzee copulation calls from the Budongo Forest. MS. with sampling gaps of no greater than two days. and assessment of fertile period To determine approximate timing of ovulation. Uganda. If more than one female was in oestrus. between Jan -Apr 2006 and Oct 2006-Mar 2007. I observed 287 copulations to which females produced calls on 104 occasions. Uganda. I determined the status of each male by calculating the proportion of other males in the community from whom he received pant grunt vocalisations.

(7) Mean Pitch: average pitch over call unit (Hz) (see figure 5. (2) Peak frequency at call beginning: frequency where maximum acoustic energy occurs at call beginning (Hz). figure 5.2) at the relevant points indicated in figure 5. (5) transition onset: frequency modulation between call beginning and middle (Hz).Chapter 5: Copulation call structure Acoustic analyses Using PRAAT software I focused on the following temporal parameters: 1) Call bout length (s): total duration of the copulation calling bout. 72 .2).1 and 5.1. (6) transition offset: frequency modulation between call middle and call end (Hz). The following frequency measurements were taken: (1) Peak fundamental frequency: location in the frequency domain where maximum acoustic energy occurred in the F0 in the call middle (Hz). (4) Peak frequency of the call at call end: frequency where maximum acoustic energy occurs at call end (Hz). Measurements of the frequencies at which maximum acoustic energy was present were obtained by creating spectral slices (Amplitude plotted against frequency. 2) N call units: number of individually distinct call units within the calling bout and 3) inter-call interval (s): time duration between each successive call unit. (3) Peak frequency of the call at call middle: frequency where maximum acoustic energy occurs at call middle (Hz).

73 . to facilitate transition onset/offset determination. middle. Peak frequency measures in the fundamental were taken initially at the beginning. The first spike (A) denotes the peak frequency of the fundamental and the second spike (B) denotes the peak frequency of the call.2).2) H: Transition onset I: Transition offset.1).2: Illustration showing how peak frequency of the fundamental (J) and of the call (E. Mean pitch is not illustrated (see appendix I for identical procedure). B: Call beginning C: Call middle D: Call end. see figure 5. middle and end of call. AB Figure 5.FG) were measured at each call stage (beginning. end. figure 5.1: Illustration of how spectral parameters were taken: A: Fundamental frequency band. though only call middle remained in the final model (J. E: Peak Frequency of call at call beginning F: Peak Frequency of call at call middle G: Peak frequency of call at call end (see figure 5.Chapter 5: Copulation call structure J Figure 5.

I analysed all calls given during a bout and calculated a median value for each parameter. They are therefore a sensitive measure of co-linearity and highlight potential problems of instability in a model. 2007). I used a permutated Discriminant Function Analysis (pDFA) to estimate the significance of the number of correctly classified calls (crossvalidated). 2003). 74 .0 (Slocombe and Zuberbuhler. I then used a Discriminant Function Analysis (DFA) to verify whether acoustic structure could be quantitatively assigned to particular variables.5. All ten acoustic parameters were used to generate the discriminant function. To determine the effect of location on calling probability I conducted a binary logistic regression (see chapter four).Chapter 5: Copulation call structure Statistical analyses I performed co-linearity checks for the 10 parameters and found that all had satisfactory variance inflation factors of less than 8. e. male rank and cycle period. it has been argued that conventional DFA does not allow for a valid estimation of the overall significance of discriminability (Mundry and Sommer. 2007). Since the acoustic data for male rank were two-factorial (female ID.05. All tests were carried out using SPSS version 15.1 (R core development team).0 and R version 2. Deschner et al. male rank). If the original DFA achieved significance. 2000. Alpha levels were set at 0. Univariate GLMs account for multiple contributions from the same individual by incorporating female ID as a random factor in the model (Pinheiro and Bates.g. Variance inflation factors (VIF) measure the degree to which the variance of one parameter is inflated by the existence of linear and higher order correlation amongst other parameters in the model. If appropriate.. which was then subjected to statistical analyses. I ran MANOVAs and two-way univariate GLMs to explore the effect of categorical variables on the acoustic structure of copulation calls.

NB: N=2. P=0. Female Ovulatory Status To investigate whether females provided reliable information about their fertility state.42)=7. P=0.001. 5 of the 10 acoustic parameters varied significantly amongst individuals throughout the cycle.42)=4. KY: N=5). 2-tailed).350.42)=3.176. Because I found no differences.167).47. P=<0.41. P=0.42)=3. and mean pitch (F(5. WL: N=2). A two-way between-subjects ANOVA (with female identity as a random factor) indicated that the female’s fertile phase had no effect on any of the 10 temporal or spectral acoustic parameters (all p-values > 0. Wilks Lambda=0. I compared the acoustic structure of copulation calls given during their POP (Ncalls=16) and Post-POP (Ncalls=10) from five complete maximally tumescent oestrus cycles (LL: N=1.001).016). P<0. WL: N=19. NB: N=9. JL: N=4.879.085. MK: N=11.431. Calls could be classified to individual females with 56% accuracy (cross-validated). suggesting that fertility was not acoustically marked in chimpanzee copulation calls. I analysed 256 calls. constituting 71 calling bouts from 6 cycling females (LL: N=23.808. P<0.001). peak frequency at call end (F(5.1 see table 5.002). A between-subjects MANOVA showed that female identity significantly affected the acoustic structure of copulation calls (F(5. P=0. or even 75 . Discriminant Function Analysis revealed that the 10 acoustic variables explained a significant amount of variation amongst the individual female callers (Wilks’ lambda = 0.42)=4. The acoustic parameters that accounted for most of the variance were all frequency measures: peak frequency in F0 (F(5.001.60)=2.Chapter 5: Copulation call structure Results Female Identity To determine if female chimpanzees gave individually distinctive copulation calls.001). χ2(55): 141.288. P <0.1). transition onset (F(5. a rate significantly above chance expectation (Binomial (0.006). peak frequency at call beginning (F(5.

76 .20) 0.316 P Value 0.055 (F 1.861 (F 1.485 (F 1.525 0.887 0. in any of the acoustic variables.54)=24.064.19) 0.54 0.624 0.435 0.14) 0.74 0.528 (F 1.20) 0.154 (F 1.4).829 (F 1.006.833 0. A between-subject univariate ANOVA. P=0.20) 0.026 (F 1. . revealed that peak frequency at call beginning was significantly higher when females mated with high-ranked than with lowranked males (Sidak corrected for multiple comparisons F(1. during POP and Post-POP.1: Results of Univariate ANOVA’s. with female identify as a random factor.Chapter 5: Copulation call structure trends. comparing each acoustic parameter measured.139 (F 1.20) 0.624 Male dominance rank To determine if female copulation calls vary acoustically with male rank. I refrained from conducting further discriminant function analyses.19) 0. see figure 5.439 0.20) 0.20) 0. I analysed calls from five females whilst copulating with high and low-ranked males. Acoustic parameter Call duration Call unit duration Interval duration Peak frequency in fundamental middle Peak frequency at call beginning Peak frequency at call middle Peak frequency at call end Transition onset Transition offset Mean pitch F-Value (F 1.18) 0. Table 5.73 0.3 and 5.313 (F 1.

Chapter 5: Copulation call structure Figure 5. Differences are mainly caused by shifts in the location of acoustic energy from the fundamental frequency to one of the harmonics at call onset. all parameter level.682. When taking all 10 acoustic parameters into account. Given that only one acoustic variable differs between male ranks. a level below that expected by chance.3: Mean peak frequency at copulation call beginning (Hz) for each female when copulating with high and low-ranked adult males.398).3% of copulation calls were correctly classified (cross validated) according to the rank of the mating male.615. Only 46. this last DFA finding was to be expected. P=0. Error bars represent mean +-1SE. 77 . χ2(12)=12. Therefore on a more crude. though at a more fine scale there are subtle and striking differences. copulation calls do not seem to differ with partner rank. a DFA failed to explain a significant amount of variation in the acoustic structure of copulation calls given with low and highranked males (Wilks’ lambda =0.

An identical analysis with high-ranking males found no such difference between tree and ground (B=0.1).148.Chapter 5: Copulation call structure (i) (ii) Figure 5.001). Finally.4: Time-frequency spectrogram for female chimpanzees (i) LL and (ii) WL when copulating with high and low-ranked adult males. Visible are the fundamental frequency (A) and the first two harmonic bands. 78 .1Hz. Filter bandwidth: 159Hz. Frequency resolution: 86.575. (B) indicates the frequency at the beginning of the call where maximum energy occurs (peak frequency of the call). their vocal behaviour was affected by location. calling females copulating with a low-ranking male called significantly more if they were exposed in a tree than on the ground in more dense vegetation (controlled for copulation number. P<0. When at least one high-ranking male was present (N=80). binomial logistic regression B=-2. in cases when females mated with low-ranking males (N=98). P>0.

the relative social position of the male mating partner.Chapter 5: Copulation call structure Discussion In the previous chapter.Whether other chimpanzees perceive these acoustic differences will have to be further investigated. 2008). young low ranking males will play an important role in the intermediate future (Boesch and Boesch Achermann. probably to confuse paternity and to gain the future support of adult community males (see chapter four). 1986. 2008a). It has been shown that chimpanzee males exhibit mating preferences for older females (Muller et al. Whilst females are probably primarily interested in copulating with high-ranking males. Pradhan et al. Stumpf and Boesch. 2006). In particular. 2000. a pattern also found in barbary macaques (Pfefferle et al. 2006). I have shown that female chimpanzees use copulation calls in flexible and potentially strategic ways.. Reynolds. My results here further support the paternity confusion hypothesis but also highlight that natural selection appears to have favoured acoustic variation within these calls. 2007). 2000. 2005. 1979. females prevent monopolisation by a socially dominant male and hereby increase paternity confusion (Pradhan and van Schaik. possibly because they require fewer cycles before conception (Deschner and Boesch. By concealing ovulation. they do not provide accurate information about the timing of ovulation... It is therefore possible that ovulation concealment has evolved under the selection pressure of paternity confusion amongst highranking and to some extent lower ranking males (Tutin. 2005). Goodall. Given that concurrently cycling oestrous females may compete with each 79 . calls may encode caller identity and very subtly. I interpret these findings as follows: Firstly. Boesch and Boesch Achermann. although chimpanzee females appear to be motivated to advertise their receptive state through calling. a strategy that is likely to secure them with future protective support.

and this was one aspect of calling behaviour that remained unexplained. when attending to calls. However. in these instances. observations). this study has also indicated that the dominance rank of the mating male has a subtle effect on the acoustic structure of primate copulation calls. keeping potential listening male mating partners informed as to the whereabouts of certain females. usually after taking active steps to ensure furtive mating (Pers. discriminant function analysis failed to separate the calls overall. Whilst it has previously been shown in yellow baboons that temporal acoustic parameters differ with the rank of the male mating partner (Semple et al. females produced copulation calls significantly more often if mating took place in a tree where the female is visually exposed and has little control over a male’s solicitation. and because of this. Finally. females sometimes do still produce copulation calls in these circumstances. what advantages could there be to copulating females signalling the rank of the mating male in their copulation calls? I found that female chimpanzees generally prefer to remain silent when copulating with lowranking males. Indeed. 2002).Chapter 5: Copulation call structure other over access to males (see chapter four) the ability to convey identity becomes a useful tool. no difference was found in frequencybased call measures. My results here suggest this may be related to the likelihood of detection by other group members.. a recent instance of attempted 80 . When mating with a low-ranking male. Only one acoustic variable out of ten measured. it is not unlikely that chimpanzees are able to use these acoustic cues. most surprisingly. high ranking males. However. varied with male rank. From a more functional standpoint. and that even very subtle changes in the acoustic structure are salient to listeners.

Furthermore. High-ranking males have been seen to attack females for copulating with low-ranking males and it is thought that such sexual coercion helps improve high-ranking male access to preferred females (Van Schaik et al. may allow females to control the potential for aggression from high-ranking males in these risky situations. Kahlenberg. females are providing some form of honest information regarding the rank of her copulating partner.Chapter 5: Copulation call structure solicitation by a male in a tree resulted in a 40m fall. 2004. In line with the results presented in chapter four. high-ranking males appear to avoid confrontations with other males. in contrast to when they are in the dense vegetation on the ground. Given that multiple females can cycle simultaneously and that with 81 . Mating with a low-ranking male is socially risky. 1999). owning up to their dealings. Matsumoto-Oda. and thus it is normally the females that receive the brunt of such male aggression (Goodall. rather than continuing to mate surreptitiously. which ultimately ended with the female’s death (BCFS long term data). and it is possible that this decreases male aggression. Hauser. 2006). Whilst high-ranking adult males represent the most efficient short-term protectors. hence mating and paternity confusion amongst these males may also be part of a female’s long term sexual strategy. these data suggest that the primary function of chimpanzee copulation calls is to spread the probability of paternity more evenly amongst adult male group members enlisting their future protective support.. I tentatively propose that. with an acoustically distinctive copulation call. low-ranking males can become highranking in the future. my analyses suggest that the identity of the female is encoded within copulation calls and so potentially available to listening males. 1986. Interestingly. even the low-ranking ones. 1990. In other words. Not only are tree matings more precarious but females are generally prevented from furtive mating.

whilst males should be interested in maximising paternity certainty (Hrdy. in addition to including more copulation calls from high and low ranked mating partners. Pradhan et al. However. 1979.. 2005) male-female aggression was rarely observed over the 10-month study period (35/287 copulation events). shaping not only call production.. and the aggression that accompanies it. aggression and habitat location need to be implemented. To test this last theory more thoroughly. these findings suggest that female competition. Females chimpanzees should be interested in paternity confusion. 1977. Taken together. long-term data analyses on calling behaviour. but also the fine acoustic structure of copulation calls. Van Schaik et al. more detailed acoustic and location analyses did to some extent: copulation calls acoustically label the rank class of the male partner. see also Stumpf and Boesch. reducing the risk of male and female infanticide. Muller et al. However. 2004).. Possibly due to the unusual male:female sex ratio at Budongo (1:3) (chapter three. 2006). It should be stressed that this is just a working hypothesis and one potential explanation of many. 2004. records from previous years with more representative sex-ratios 82 . Although my results on copulation calling frequency (chapter four) did not seem to indicate this. it would be naïve to think that female competition is the only relevant social pressure female chimpanzees are exposed to in the reproductive context. may be more of a signifant social pressure than previously recognised.Chapter 5: Copulation call structure increasing age some females become preferred over others (Tutin.. Female and male sexual strategies rarely complement each other and strong conflict of interest is often the result (Van Schaik et al. 2006) labelling one’s own identity may increase overall fitness levels. Honestly indicating the rank of the mating partner may be an additional vocal strategy employed by female chimpanzees to moderate the aggressive potential of nearby high-ranking males.

) However. but because of their ubiquitous occurrence. To date. 2005. Victim screams therefore provide the perfect oppurtunity to test how females chimpanzees comprehend vocalisations and what potential social benefit this affords them. Pers. are less likely to adversely affect the future success of these animals. The logical next step will be to experimentally verify whether these calls are meaningful in this way to other group members and to what degree males. no close range playback has been conducted with wild chimpanzees to test what they understand from calls and how they can use this information to their social advantage. Because chimpanzee copulation calls are likely to play a crucial role in a female’s reproductive success it would be unwise to attempt the first comprehension playback with such an important vocal signal. Like copulation calls. In chapter six I sought to overcome this problem by experimentally testing the meaning of victim screams. Newton-Fisher. or indeed females. are guided in their behavioural decisions by these acoustic cues.Chapter 5: Copulation call structure indicate that this is not the norm and female-directed male aggression does indeed occur (BCFS long term data. until such data are available. 83 . these results should be viewed cautiously. comm. these vocalisations play an important role in the social lives of wild chimpanzees.

although they conveyed very different information. Subjects consistently discriminated between the two scream types in ways that suggested they assigned meaning to these calls. The results from this chapter have been submitted for publication Slocombe. These agonistic screams vary acoustically as a function of the severity of aggression the caller is facing. I compared subjects’ responses to screams given in severe and mild aggression. I therefore suggest that by attending to conspecifics’ calls. Townsend S. & Zuberbuhler. Animal Cognition. Wild chimpanzees distinguish between different scream types: evidence from a playback study. Possibly due to the challenging nature of their vocal repertoires and social systems. I conducted a series of field playback experiments with a community of wild chimpanzees in the Budongo Forest. Uganda. Acoustically. To determine whether individuals could meaningfully distinguish between screams given in different agonistic contexts.. these results demonstrate that chimpanzees are able to meaningfully distinguish between these graded scream variants. I employed a control condition. In order to control for the possibility that response differences were a product of the screams’ peripheral acoustic features. In this study. tantrum screams matched severe agonistic screams in their peripheral acoustic features.W. no previous attempts have been made to experimentally test call comprehension in any wild great ape species. 84 . Accordingly subjects showed little interest in these calls.E. rather than any encoded social meaning. K. I used tantrum screams.Chapter 6: Scream comprehension Chapter Six: Wild chimpanzees distinguish between different scream types: evidence from a playback study Summary When experiencing aggression from group members. chimpanzees can understand the nature of social events they cannot see and potentially use this information to make future evasive behavioural decisions. chimpanzee females commonly produce screams. but no physical threat. given by individuals that experienced extreme social frustration. Together. Recent research has indicated that females can compete aggressively and even lethally for resources. K. (accepted).

Chapter 6: Scream comprehension Introduction Discriminating between signallers and receivers and investigating the associated behavioural and cognitive processes separately, is one particularly useful approach to studying animal communication (Seyfarth and Cheney, 2003). In the previous chapters I have focused on copulation calling in wild chimpanzees and how the use and acoustic structure of this vocalisation has been potentially shaped by social pressures. This work has begun to address the question of chimpanzee vocal counter-strategies to female competition, but primarily from the producer’s perspective. If we are to get the full picture as to the role vocalisations play in mediating female competition, it is necessary to understand what the receivers can extract from vocalisations. In this chapter I attempt to address this question for the first time in any wild ape species, focusing specifically on what female chimpanzees understand from agonistic screams.

Once considered a solely male affair, intense aggression is now known to be an important feature of both male and female chimpanzee social lives (Kahlenberg et al., in press (a); Pusey et al., in press; see also chapter three). In females, such aggression is thought to be a result of competition for resources such as food, or even male mating partners and can have considerable fitness consequences, sometimes culminating in lethal infanticidal attacks (chapter three).

Due to the dense nature of the chimpanzee natural habitat and the fission-fusion nature of their social system however, female chimpanzees cannot always directly observe ongoing conflicts. Therefore, the ability to make references about the nature of events would be clearly adaptive, allowing individuals to make a number of important decisions, such as whether or not to intervene, or to avoid the ongoing interaction. Furthermore, the ability to monitor out-


Chapter 6: Scream comprehension of-sight agonistic interactions will greatly increase the amount of social knowledge about third party relationships an individual female can accumulate. Compiling of social knowledge is likely to have direct fitness benefits and may have even driven the evolution of large brains in primates (Humphrey, 1976).

Previous research has indicated that non-human primates appear to be capable of extracting meaning from calls given during agonistic interactions. The agonistic screams of rhesus macaque monkeys for example, have been shown to vary with the severity of the attack and the rank of the opponent (Gouzoules et a., 1984). Subsequent playbacks showed mother macaques were sensitive to this information conveyed in their offspring’s screams. Females responded most to screams consistent with high risk of physical harm to the infant or a threat to the matrilineal dominance by a lower ranking individual (Gouzoules et al., 1984).

Similarly to macaque monkeys, chimpanzees also produce context-specific screams. Slocombe and Zuberbuhler (2007) have recently shown victim screams can vary with the severity of attack an individual is experiencing (reviewed in chapter one). However, whether listening individuals can meaningfully differentiate between these graded call types still remains unknown.

In collaboration with Katie Slocombe, I conducted a playback experiment to test if listeners could distinguish between victim screams elicited by different types of aggression. We conducted the study with a habituated group of wild chimpanzees in the Budongo Forest, Uganda. To date, only two previous studies have used playbacks with wild communities. As discussed in chapter one, in both cases the focus was on how chimpanzees responded to the presence of extra-community individuals. If we are to understand subtle social cognition, it is


Chapter 6: Scream comprehension crucial to playback calls from recognisable group members. Numerous studies with freeranging baboons have already provided vital insights into the meaning and function of intragroup social signals (e.g. Bergman et al., 2003; Wittig et al., 2007), but comparable efforts have never been attempted with any wild ape species.

This lack of effort in any of the great apes is most likely due to two factors. To begin with, most great apes, including chimpanzees, have extremely graded vocal systems that are inherently difficult to examine (Marler and Tenaza, 1977). Secondly, in general, fieldworkers have refrained from applying experimental techniques to wild apes, despite the well-known power of playback experiments as tools in the investigation of call comprehension (Seyfarth et al., 1980). So far, only a single captive playback experiment has been performed with great apes to probe their understanding of conspecific calls (Slocombe and Zuberbühler, 2005b),

An additional aim of this final chapter therefore is to devise an ethically and methodologically sound protocol for conducting playbacks with free-ranging chimpanzees.

I predicted that if chimpanzees distinguished between different types of agonistic screams, they would show more interest in screams linked with significant social consequences. That is, cases in which the aggression was most severe and thus potentially likely to pose a threat to surrounding chimpanzees, what is sometimes termed “snowballing”. In line with previous research (e.g. Cheney et al., 1995; Gouzoules et al., 1984) I expected the chimpanzees to demonstrate this with a longer looking duration in the direction of the calls and a faster latency to orientate towards them. Several recent monkey playback studies have simply compared responses to the acoustically distinct signals of interest (e.g. Pfefferle et al., 2008b), however, this leaves the possibility that subject responses are being driven purely by the low


they are even more intense signals than the screams given during conflicts. if chimpanzees were extracting information about the likely eliciting event from the call. where we presented subjects with tantrum screams given by group members. in the Budongo Forest. but with varying degrees of habituation (for more details see chapter two). To control for this possibility we included a control condition. if they are responding to the most salient and intense vocal signals in their environment. Alternatively. with no regard for the underlying informational content. Tantrum screams are approximately matched in acoustic structure to the severe screams. all of which were individually known. During the period of study (Jan-November 2007) the group comprised 78 individuals including 8 adult males. 2005). but. and a less so to mild victim and tantrum screams. then tantrum screams should be fairly irrelevant to them. differential responses to these calls could be directly driven by the saliency and the physical features of the two call variants (Owren and Rendell. 25 adult females. as screams given to severe aggression are longer in duration and higher in pitch than screams given to mild aggression (Slocombe and Zuberbuhler. they should react most to severe victim screams. In this case. Methods Study area and Subjects The playback study was conducted on the Sonso community of chimpanzees. Uganda (Reynolds. 1997). to the human ear. 2007).Chapter 6: Scream comprehension level acoustic features of the calls. 6 sub-adult males and 3 sub-adult females. To summarise. 88 . despite their overt acoustic features. they should show the greatest reaction to tantrum and severe screams and a weaker response to mild screams. If subjects processed the informational content or meaning of the screams.

Owren. Subsequently. mean pitch. mild and tantrum.37 and scripts written by M. Chimpanzees of this community are known to react to screams occurring within about 30m by looking towards the sound source and sometimes by becoming pilo-erect. as their suspicions would have been aroused upon finding no chimpanzee at the corresponding location. see table 6.1). I then calculated a median value of each acoustic parameter for each stimulus. displaying or approaching. Pilot observations (conducted by Katie Slocombe) indicated this duration was long enough to elicit interest in the stimulus without inciting approach. As reactions could also be affected by relatedness of subject to stimuli-provider I only played screams of unrelated individuals to subjects.Chapter 6: Scream comprehension Playback stimuli Playback stimuli consisted of victim screams given in response to severe and mild aggression. who provided several exemplars of screams to both severe and mild aggression. mean frequency of the first formant and peak frequency in the fundamental. see appendix I). This acoustic analysis was performed using Praat version 4. 89 . I was able to confirm the pattern of acoustic differences between mild and severe victim screams reported by Slocombe and Zuberbühler (2007) by measuring each call in each stimulus along four parameters (duration. We therefore limited the playback stimulus to a short bout of screaming of approximately 3 seconds. Sub-adult males are generally considered as more innocuous members of the group and therefore their screams provided the perfect way to test call comprehension without inducing high arousal responses. We did not want the chimpanzees to approach the source of the noise. I averaged the values for the stimuli in each stimulus class (severe. The stimuli came from five sub-adult males.3.

31 10 (p=0.429) Calls used as playback stimuli were recorded opportunistically from known individuals by KS and ST using a Sennheiser ME66 microphone and a Sony TCD-D8 portable DAT recorder (KS) or a MARANTZ PMD 660 solid-state recorder (ST). using Cool Edit Pro LE. N refers to the number of scream bouts (used as stimuli) that contributed to the mean values reported in each category.4 1574 1436 SD Duration (s) 103. onto a PC (Toshiba: Celeron.429) 215 1064 10 (p=0.662) 189. Stimuli contained only one chimpanzee calling and relatively low levels of background noise (cicadas.9 62 0.Chapter 6: Scream comprehension To control for responses being driven by the saliency of the acoustic features of the signal. 16 bits accuracy. provided by two infants.6 201 1261 12 (p=0. I also tested subjects’ responses to tantrum screams. Recordings of vocalizations were transferred directly from the solid state memory card or digitally from the DAT recorder.44 0.1: The mean values and SD of 4 acoustic measures for each scream condition.2) was used to cut the stimuli and to fade the background noise in and out at the start and end of each stimulus. which were comparable in acoustic fine structure with that of severe aggression screams (see table 6. Calls were digitised at a sampling rate of 44.54 SD 0.8 38. Fujitsu Siemens: Amilo).052) 0.17 Mean Pitch (hz) 1450 1404 SD 113 58. Mann Whitney U tests revealed no significant differences between the acoustic measurements of the tantrum screams and severe screams. to avoid the playback having an abrupt onset and offset. Peak freq of 1st Formant fundamental SD (hz) (hz) Tantrum (N=5) Severe (N=6) Mild (N=7) Mann Whitney U value 1451 1435 92. Raven (Version 1.1 kHz.2). We checked that all stimuli sounded natural to an 90 .1) Table 6. wind).15 0.3 1276 4 (p=0.1 0. Each playback stimulus was approximately equal in bout length (3 seconds) and I used Raven 1.2 to equate the stimuli in terms of RMS amplitude (see table 6.

concealed in a rucksack. but had some additional features to maximise the suitability of this paradigm for this fission-fusion great 91 .1: Photograph illustrating the speaker used to conduct playbacks.04 (0. mild scream) and control (tantrum scream) trials that followed a within subject design.20) 8423 (239) Mild (N=7) 3.Chapter 6: Scream comprehension experienced human listener at a distance of 30m and could not be heard clearly behind the speaker at 100m. Table 6.01 (0..11) 8367 (160) Tantrum (N=5) 2. 1980). Stimuli were stored and played in WAV format on an Ipod Nano and broadcast using a NAGRA DSM speaker/amplifier. The speaker was concealed in a modified rucksack throughout the experimental period so the chimpanzees never saw the equipment in operation (see figure 6.2: The mean duration and RMS amplitude of stimuli in each condition. The protocol followed the same principles as traditional playback experiments conducted with monkey species (Seyfarth et al.1) Figure 6. Mean duration Mean RMS Amplitude Severe (N=6) 3.21) 8433 (325) Protocol The experimental phase consisted of test (severe scream.90 (0. with each subject receiving all three trial types in a counterbalanced order.

In addition. (3) Operators with the speaker recorded all individuals within a 50m vicinity of the speaker at the time of playback.g. Each operator collected data on a number of behaviours prior to and following a playback. one to stay with the ‘scream provider’ (the individual whose calls we played from the speaker) and one to play the stimulus from the speaker. feeding) and the time at which it changed was also recorded. To prevent the scream provider from hearing his own vocalisation he. specifically the number of aggression bouts the subject was involved in or witnessed and the duration and number of screaming bouts the subject heard. We used Motorola GP340 radios or Nokia 2600 mobile phones to maintain contact between all three operators throughout the duration of the experiment.2). or she.Chapter 6: Scream comprehension ape species. recorded all occurrence aggression. the general behaviour of the scream provider (e. the distance and 92 . All observations of grooming and pant-grunting (a social greeting vocalisation given by subordinates to dominants) received and given by the subject were also recorded. (1) ‘Focal’ operators. resting. grooming. A minimum of three operators were required to run the experiment in a fashion that maximised the realism of the playbacks: one to stay with the experimental subject – the receiver of the vocalisation (always ST). In order to maximise realism. had to be >100m away from the speaker (see figure 6. Once a playback was completed operators noted the identity of all individuals in the party to ensure that future playback subjects would not receive a stimulus they may have previously overheard. (2) Operators with the scream provider recorded the time and nature of all loud vocalisations made by members of the scream provider’s party and the identity of the vocaliser. who followed the experimental subject. They particularly looked for behavioural changes at the time of the playback.

the speaker operator took the loudspeaker 30m away from the subject in the direction of the scream provider and concealed himself. I meticulously ensured no other chimpanzees were within 30m of the speaker before the screams were played back. which may distract him from the stimulus or to make his response ambiguous) and alone (to exclude the possibility that any subsequent response to the stimulus was a result of social referencing). As soon as the subject was resting on the ground (without feeding or grooming. Figure 6. In the absence of the scream 93 . In order to prevent unwillingly exposing any chimpanzee to an intense stimulus. It was therefore vital that the scream provider did not vocalise within 10 minutes prior to the playback.2: Schematic representation of experimental set-up.Chapter 6: Scream comprehension direction of the playback had to be plausible. announcing his true presence in a different location of the forest to the subsequent playback.

I also ensured that no chimpanzee heard the same stimulus twice. the focal operator usually had to be within 10m of the subject and only the very well-habituated individuals of this community would tolerate this.Chapter 6: Scream comprehension provider’s vocalisations this experimental set-up realistically simulated the presence of another chimpanzee 30m away. When this condition was met the focal operator used a Panasonic NV-GS 250 digital video camera to film the subject for 30 seconds before and 1 minute after playback. I played a different exemplar to that individual. If a subject had been in the vicinity of a playback I conducted with another focal. Focal subjects were then followed for as long as possible after each trial and data collection continued. unpublished data). involved in either an agonistic interaction of severe or mild intensity or a tantrum. I also waited at least 1 day in between trials and the same focal subject never appeared in consecutive experiments. Another prerequisite was that the subject had to be facing away from the speaker so any orientation response was unambiguous. Range = 0-7/month). The focal operator was responsible for checking with the other operators that all conditions had been met before telling the speaker operator to play the stimulus. 94 . when alone on the ground. however playbacks were conducted at a very low rate over 10 months (mean =3 /month. Due to low visibility at this site. To avoid order effects I counterbalanced trials. rather than repeat the same stimulus they may have overheard. Screams occur naturally at a high daily rate (approximately 6 scream bouts/hour) in this chimpanzee community (Townsend and Slocombe. ensuring subjects did not hear conditions in the same order.

the subject’s reaction was consistent with the predicted direction and in the other trial the reaction went in the opposite direction to that predicted. 1990) and thus substituting one sex for another should not detract from the overall aims of this experiment. To avoid pseudoreplication I averaged her responses for each test condition. Unfortunately only four females in the Sonso community would tolerate the degree of following needed to conduct these experiments. Males and females are unlikely to differ in their call comprehension skills (Cheney and Seyfarth. Two further females only received the control condition and were not included in the analyses. mild N=2) and she showed the same pattern of response to both sets of stimuli. In one trial. However. their responses to this condition were very similar to the responses of the other six subjects tested. Two trials were excluded due to problems with the experimental set-up confounding the subject’s response. Five subjects were tested only once for each condition. subjects were tested in all three conditions. 1998). very well habituated. The following three behavioural responses were measured: (1) Latency: Time from stimulus onset to the subject orientating its head towards the speaker. Because with within-subject designs it is virtually impossible to find significant effects with samples smaller than six (Mundry and Fischer.Chapter 6: Scream comprehension Playback trials and data analysis A total of six. I had to supplement these adult females with two adult male subjects from the community. (3) Number of looks: Number of consecutive looks towards the speaker in the minute after the playback 95 . One subject heard stimuli from two different individuals (severe N=2. I coded videotapes frame by frame using Adobe Premier Pro C53 software. (2) Looking duration: Time for which the subject orientated its head towards the speaker in the minute after the playback.

was asked to analyse 25% of trials (5 trials).12hrs). We compared the measures taken by each coder and found very high levels of agreement (Pearson’s correlation for latency = 0.001).Chapter 6: Scream comprehension To ensure accurate coding of videotapes. Statistical analysis Non-parametric tests were used for comparisons (SPSS version 12. Behavioural responses Severe versus mild screams I predicted that if subjects processed the information conveyed by screams. he was required to be a minimum distance of 100m from the speaker. p<0. blind to the trial type. however during some playbacks the scream provider could be further.08. duration = 0. as opposed to asymptotic ones. a second coder (KS). SD +/.974 (2-tailed. My prediction was supported: during the 96 . as recommended by Mundry and Fischer (1998).30). As we were dealing with low sample sizes. Results Playbacks occurred between 8am and 4pm. SD+/-262).001). they should respond more strongly to severe than mild screams. Prior to playbacks subjects were followed for on average 188 minutes (SD+/131) and after playbacks I continued to follow subjects for a mean duration of 69 minutes (SD+/.2. All tests were twotailed. up to 1000m away (Mean scream provider distance = 355. p<0. number of looks = 1. typically at around 11am (Meantime of day =11. which indicates the trials had been accurately coded. we calculated exact p-values.0).93 (2-tailed. To prevent the scream provider from hearing his own vocalisation.00 (2-tailed.002). p=0.

Chapter 6: Scream comprehension minute following playback subjects looked towards the speaker for longer in response to severe screams than mild screams (Exact Wilcoxon matched-pairs-signed-rank-test: Z=-2.20, N=6, P=0.031; see table 6.3, figure 6.3). There was also a trend to look at the speaker more times and more quickly in response to severe screams than mild screams (number of looks: Z=-2.06, N=6, P=0.06; latency to look: Z=-1.57, N=6, P=0.156; see table 6.3; figure 6.3 and figure 6.4).

Table 6.3: Behavioural responses of subjects to each of the three playback conditions. Mean values across the 6 subjects and the standard deviations are shown. Severe (n=6) Mean Latency (s) 1.19 (0.35) Mild (n=6) Tantrum (n=6)

1.72 (0.72) 2.64 (0.91)

Mean Duration of Looking (s) 14.66 (6.87) 6.99 (4.12) 6.79 (5.11) Mean Number of Looks 2.58 (1.20) 1.16 (0.41) 1 (0.63)

Figure 6.3: Boxplots illustrating the total duration spent looking at the speaker and the total number of looks in the minute after playback in response to three types of playback stimuli. Boxplots illustrate medians, interquartile ranges and highest and lowest values.


Chapter 6: Scream comprehension Severe screams versus tantrum screams To address the more parsimonious explanation that subjects were simply responding to the more intense acoustic features of the severe screams I compared subjects’ responses to screams given in severe conflicts with tantrum screams. If responses were driven by the saliency of the sound and its ‘attention grabbing’ acoustic features, I would expect subjects to show roughly equal interest in both the severe and tantrum screams. In contrast to this prediction, during the minute following playback subjects looked towards the speaker for longer (Exact Wilcoxon matched-pairs-signed-rank-test: Z= -2.23, N=6, P=0.031; see table 6.3, figure 6.3), more times (Z= -2.20, N=6, P=0.031; see table 6.3, figure 6.3) and more quickly (Z= -2.20, N=6, P=0.031; see table 6.3 and figure 6.4) in response to severe compared to tantrum screams. In one control trial, the subject (NB) did not orientate to the speaker at all. I allocated her a “ceiling latency” which was the maximum latency observed from the analysed trials to allow her inclusion in latency analyses.

Figure 6.4: Boxplots illustrating the latency to look in response to playbacks in each condition. Boxplots illustrate medians, inter-quartile ranges and highest and lowest values.


Chapter 6: Scream comprehension Effect of prior exposure to screams As aggression and screaming also occurred naturally during the study period, a subject’s responses could have been modulated by these events, e.g. by how much aggression the subject had perceived prior to playback.

In order to examine the effect of prior exposure to naturally occurring screams I calculated the number of screaming bouts focal subjects could have seen or heard in the hours before a playback trial. Mean duration of screaming witnessed and heard did not differ significantly between conditions (severe versus mild: Mann Whitney exact: U=22, Z=-0.319, Nsevere=7, Nmild=7, P=0.805; severe versus tantrum: Mann Whitney exact U=18, Z=-0.429, Nsevere=7, Ntantrum=6, P=0.731, see table 6.4). This suggests that previous exposure to conflict did not explain any of the observed response differences.

Table 6.4: The mean and SD of screaming bout duration witnessed and heard, before playback, in each condition. Severe (N=7) Mean duration of screaming bouts witnessed and heard (s) 168 (189) 189 (227) 232 (195) Mild (N=7) Tantrum (N=6)

Short-term behavioural changes in response to playback experiments Playback experiments with wild chimpanzees raise some ethical concerns. For example, the possibility exists that simulating agonistic encounters with playbacks artificially increases subsequent levels of aggression in the community. To ensure that never took place, I performed a number of additional post-hoc analyses. Firstly, I investigated whether our


Taken together. although this trend was not significant (Mean scream durationhour before = 46.7).039.7 sec (SD= 66. 7 occurred within one month (January 2007) when conditions were optimal. In fact. mean scream bouts hour after =1. I found no evidence to support this prediction. I predicted that if playbacks intensified levels of aggression. I looked for longer-term effects by comparing aggression levels across years. Of the 20 playbacks considered.073.4 (SD= 5.8 sec (SD= 112. Wilcoxon test: Z=-1. mean scream bouts hour before =4.7).8 sec (SD= 29. these results indicate that playbacks had no immediate effects on the daily pattern of aggressive incidents in this community. Wilcoxon test: Z=-1.299).134). p=0. but the pattern of aggression was stable across days with and without playbacks. As aggression levels vary throughout the day. mean scream duration hour after = 44. N=20. mean scream bouts hour after =2. p=0.5). mean scream bouts hour before =2. N=20. N=20. P=0.0). Wilcoxon test: Z=-1. Long-term effects of playback experiments In a final analysis. N=20.498. I found a similar trend for there to be less aggression later in the day on non playback days (Mean scream durationhour before = 70.0).7 (SD=2.067.9 (SD=3.Chapter 6: Scream comprehension playbacks caused any immediate effects by analysing levels of aggression in the hour preceding and following playbacks.9). Wilcoxon signed rank test: Z=-1.6). mean scream duration hour after = 18. p=0. It is apparent there was generally less aggression on days when I performed a playback experiment (probably because it was more difficult to meet all conditions necessary to perform a playback on days when aggression levels were high).829.790. then more aggression should be observed in the hour following experiments.7 sec (SD= 61.95 (SD= 2. I compared all occurrences of aggressive interactions that took place 100 . I found a trend towards more aggression occurring in the hour before playback compared to the hour after. I also analysed exactly the same time periods on ‘matching control’ days when no playback occurred.4). with less fighting generally occurring in the heat of the day.

93 aggression bouts per hour (73 bouts observed over 79 hours). subjects looked at the speaker for longer and tended to look more times. but my impression was that playbacks did not appear to influence the usage of these behaviours. I found no evidence for the hypothesis that responses were driven directly 101 . Finally. This indicated that. I recorded grooming and pant-grunt vocalisations prior to and after playbacks. the scream provider was never observed to show any orientation or approach to the speaker or a change in broad behavioural category (rest. feeding) at the time of playback. as demanded by the experimental protocol. and with shorter latency than to the same individual’s screams given to a case of mild aggression. that is. In addition. the individuals whose screams I played back could not hear them and were not adversely affected by playbacks. there were on average 0. to see if other social behaviours were affected.Chapter 6: Scream comprehension in January 2007 and compared them with the matched period in the previous year (Slocombe. In January 2006. This shows that chimpanzees are capable of extracting information about the severity of the attack from the scream.97 aggression bouts per hour (48 bouts observed over 49 hours focal time) and in January 2007. they appeared to process the calls’ meaning. Discussion This study has found that chimpanzees show more interest in screams given during severe rather than mild aggression. When listening to a short bout of screaming from an unrelated individual in response to severe aggression. travel. Grooming and pant-grunting only occurred once before a playback and twice following a playback. Based on these data. unpublished data). I therefore could not conduct any meaningful analyses. there were on average 0. I concluded that scream playbacks did not adversely affect general levels of aggression in the Sonso community.

for less time and less often to tantrum compared to severe screams. and a low ranking young male (MS). most especially if the aggression escalates. Whilst this is a potential drawback. they were the most female-like in their behaviour. Whilst tantrum screams matched or even surpassed severe victim screams in a number of acoustic features. showing they were not simply orienting to the most acoustically salient sounds. Chimpanzees looked at the speaker more slowly. These were a well-established. only four females could be tested and I had to supplement these females with two male subjects. Of all the males. The aim of this chapter was to determine what information female chimpanzees can extract from victim screams. As explained earlier. foraging alone and generally avoiding male-male conflict (Townsend. snowballing out of control (Goodall. Tantrum screams are largely irrelevant to most community members and the lack of response indicated subjects were processing these calls. potentially at a semantic level. 1986). My study is consistent with the idea that chimpanzees attend to the meaning of screams and can infer the severity of an attack from listening to the screams alone. males and females are unlikely to differ in their comprehension skills of socially important acoustic signals (Cheney and Seyfarth. Pers. The potential risk to the listener can be extrapolated from the type of aggression being experienced by the caller and therefore it makes sense for resting individuals to attend more to events that may signal greater danger for them. spending time away from the core main group. though older male (MA). obsv).Chapter 6: Scream comprehension by the acoustic features or physical saliency of the stimulus presented to them. As a consequence evasive action may be needed. When a female is resting on the ground she is vulnerable to attack. Ultimately females may be 102 . individuals showed little interest in this call type. 1990).

this is critical to obtaining meaningful results and minimising the chances of disturbing the animals by presenting them with implausible or impossible events. consistent with the notion chimpanzees process the meaning of the call. they are essential to generate a realistic scenario with animals of heightened social awareness.g. By attending to these screams. such as the chimpanzee. raising issues of ecological validity. Cheney et al. I found that individuals distinguish between different types of screams.Chapter 6: Scream comprehension able to use these screams to make informed future behavioural decisions that increase their overall fitness. As with all playback experiments. For ethical reasons. They also extract the corresponding social information from the screams they hear.. 1995. Field playbacks using social signals have led to great advances in our understanding of what information calls convey and the social-cognitive skills of other primate species (e. I consider it essential that measures of aggression and other social behaviours continue to be collected ensuring that these field experiments do not negatively impact on the natural behaviour of these animals. with a correspondingly slow rate of data collection. The current study has demonstrated that field playbacks are a feasible method for work with wild chimpanzees. Moreover the customised protocol outlined will hopefully stimulate further research on chimpanzee vocal communication and social cognition. Bergman et al. which form part of a larger graded call system. To conclude. Up until now the only work to experimentally address call comprehension in any ape species has been performed in captivity (Slocombe and Zuberbuhler 2005b). Although the number of protocol prerequisites is onerous. rather than just responding to salient acoustic features.. this study has demonstrated that intra-group playback experiments using social signals can be successfully conducted with wild chimpanzees. 2003). females can potentially 103 .

104 . Here I demonstrated that vocalisations do play a crucial role in the lives of female chimpanzees. the signaller constitutes only one half of the story. In the previous chapters.Chapter 6: Scream comprehension avoid escalating female aggression and hence this may be one viable strategy employed by female chimpanzees to dissipate the risks associated with their competitive social lives. However. I have focused more on call production. shedding light on such communication from the receiver’s perspective. specifically how females use their vocalisations in potentially tactical ways.

105 . I will conclude by looking further ahead. suggesting potential future studies to address this topic more systematically. bringing them together to address the original questions.Chapter 7: General discussion Chapter Seven: General Discussion Summary In this chapter I will reflect on the findings from the previous empirical chapters. particularly: how significant a threat is competition to wild female chimpanzees and how can they use their vocalisations to mediate this social pressure? I will also focus on what further information these results can provide us with for understanding chimpanzee vocal communication and the cognitive processes that underpin it.

This crude stereotyping has resulted in the assumption that females have to deal with fewer social pressures than males. Through aggression. In comparison to the competitive. tactical and lethally aggressive male chimpanzee. leading many to assume female competition was not a significant social process in the lives of 106 . Direct contest competition pervades the lives of many female primates. particularly the chimpanzee. but rank is transient and can be altered through a variety of strategies such as aggression or tactical affiliation (Goodall. They reside in fluid social groups. Whilst. Males dominate one another. Males are also the philopatic sex and cooperate together to defend territories against incursions by hostile neighbouring communities. and this has often led to the female sex becoming neglected. particularly the more competitive one (reviewed in Hrdy. a great deal of our knowledge regarding chimpanzee social complexity stems from research on male chimpanzees. it is unlikely that the female chimpanzee’s social world is more easily navigable than the male’s.Chapter 7: General discussion Female competition in chimpanzees – a social threat? From over 50 years of wild research. there are inevitable differences between the sexes. of course. 1986. 1981). Newton-Fisher. appeared to conform to their passive role. philopatry and female alliances. For a while. it is now known that chimpanzees lead complex social lives. female great apes. the size and composition of which are constantly changing. male mammals have always been portrayed as the more socially intricate of the two sexes. dominants consistently outperform subordinates. As is clearly evident. females are often depicted as peaceful and passive individuals. 1997). in evolutionary studies of animal behaviour. Historically. A growing body of research over the last ten years on female competition is finally beginning to dispel this commonly held myth. attentive purely to the upbringing of offspring.

1993..Chapter 7: General discussion chimpanzees (Wrangham. primarily because it is more easily quantifiable (Wrangham. Between 2001-2006. Bates. Emery Thompson 107 . A similar relationship has been documented in the Kanyawara community of chimpanzees. Uganda. Exactly what the females in the Sonso community are competing over is not completely clear. Such a high influx of new females is likely to have put pressure on community resources and females responded aggressively to this increased demand. 1997). mating partners and their sperm can also become a resource that is competed over (Bergland et al. 2002). Rosvall.. At Sonso. the main focus is on competition over food. Over a period of 28-months three instances of lethal coalitionary female aggression in the form of infanticide occurred in the Sonso community. Koenig. 2001. in Budongo chimpanzees (chapter three). 2005. 2008). I provided further evidence for competition amongst females. Although the community is known to have a high food density relative to other studied communities (Newton-Fisher. These attacks simultaneously coincided with a dramatic demographic shift in the community. thereby increasing the number of resident females and skewing the sex ratio. food competition is one obvious explanation. or where males differ in their quality. 13 new females immigrated.. with females also showing heightened levels of aggression during times of immigration. (1997) challenged the notion of the passive female pawn. showing that females vary in their reproductive success and highlighting contest competition for resources as a potential mechanism driving this variance. In primate groups where the male:female sex ratio is skewed. 2003. In most studies of female competition in primates. 1980.’s (1997) original hypothesis: competition for resources can have a significant contest component. Pusey et al.. However through combining behavioural data from multiple years. Palombit et al. These observations therefore provide crucial evidence in support of Pusey et al.

the sperm of males may be one resource females are interested in securing. it is very plausible that residents are also competing over the males for the potential post-natal services they can offer (Palombit et al. many dismissed these cases as “pathological” behaviours because of the relatedness 108 .Chapter 7: General discussion et al. and that infanticide represents a significant threat to female fitness. 2005) and male sperm quantity is known to decrease with consecutive ejaculations. Given that chimpanzee males differ in their ability to provide a sufficient protective service for females and their offspring. It should be noted that these female-led infanticides are not the first to be documented in chimpanzees. probably due to their previous mating history (chapter three). This may be particularly true at Sonso where. Palombit. due to the death of the previous alpha male (DN). High-ranking males are also the most efficient protectors and perpertrators of male initiated infanticide. 1997. 2001. however. 2000).. the high rate of female immigration. there remain only four highranking males. 2007). 2006). there were almost three females for every male. Because multiple promiscuous females can cycle at any one time (Deschner et al. which is true for a number of primate species (Van Schaik. combined with too few adult males to expand the community home range. In this community. I showed that high-ranking males intervened during the female-led infanticides. 2000. Over the14 month study period. 2003. Stumpf and Boesch. means that food availability may become limiting amongst resident females.. such a highly female skewed sex ratio also highlights males as a resource worth competing over. Goodall (1986) also observed such female-initiated attacks in the 1970’s. Additionally.. Pradhan et al..

this does not necessarily translate into social simplicity. 1986. 109 . carry substantial fitness consequences. at times. A further important outcome of these observations is therefore that males and females may not differ in their aggressive propensities and tactics as much as previously thought: lethal coalitionary aggression is not necessarily a gender specific trait in apes. my findings provide clear evidence that under certain socio-ecological conditions female chimpanzees are subject to significant social pressures and threats that. Together these observations suggest that lethal female aggression is probably part of the wider female behavioural repertoire. In these instances at Sonso. in press). In conjunction with research from other chimpanzee field sites (Kahlenberg. a behaviour previously thought only to be employed by males (Goodall. at present. females do lead more subtle and secluded lives than males (Goodall.b. Pusey et al. particularly primates. we know very little regarding how females cope with these demands. Emery Thompson et al. Vocalisations are known to function as crucial tools in the social lives of many mammals.Chapter 7: General discussion of the female perpetrators (mother and daughter dyad) and the observations that they always consumed the victims. competitive and potentially dangerous is a relatively new concept.. 1997. In chapters 4-6. multiple females were involved and the infant bodies were never seen to be consumed. Calls as social tools The idea that the social fabric of female chimpanzee lives is complex. in press. They also suggest that females are capable of working together and forming aggressive coalitions against other females. in press a. in general. My results do not support this explanation. Because of this. 1986). Whilst. Wilson and Wrangham. I specifically investigated this gap in understanding and proposed that vocalisations play an integral role in mediating female competition and the aggression that accompanies it.. 2003).

Females were calling more in situations when they could advertise their receptivity to more high-ranking males. 1993. To this end. particularly the threat of female social conflict. Thus. seemed to better explain call production in female chimpanzees. both production and withholding of the behaviour needs to be understood. This was primarily driven by the observations that chimpanzees call at much lower rates than other primate species (Pradhan et al. I found that females were probably calling to confuse paternity amongst the high-ranking males of the community. there seemed to be an advantage to the strategic use or comprehension of these vocalisations. it seemed that not calling was the default behaviour. I analysed calling behaviour from a dual perspective: why females produced copulation calls and importantly. why they did not. 2006). if we are to fully understand the function of copulation calling. To address how females may use vocalisations to deal with female competition. but at the same time they were concealing their proximity to ovulation. 2005). Copulation calls In chapter four. McComb and Semple.Chapter 7: General discussion One commonly held theory suggests that vocalisations facilitated the evolution of complex primate social living (Dunbar. I found that social factors. In the face of female competition and its accompanying aggression.. Fertility concealment has been addressed both empirically and 110 . I showed that most traditional explanations cannot be invoked to explain why female chimpanzees produce copulation calls. I chose to focus on more commonly produced and evolutionarily relevant vocalisations: copulation calls and victim screams. Instead. Furthermore. an often neglected aspect in previous studies.

Male infanticide is a particularly pervasive feature of baboon life (Palombit et al. 2001. 2008a) and is generally considered as a strategy to enhance paternity confusion amongst males. similar behaviours have been documented. 2007).Chapter 7: General discussion theoretically (Semple. Although protection from male infanticide has been 111 . 2004. As previously described. In at least one of these three instances. and represent a real threat to fitness for any female chimpanzee. Van Schaik et al. high-ranking male chimpanzees were directly observed trying to prevent the ensuing attack. Pfefferle et al. being involved in three out of five documented instances. Van Schaik et al. At Gombe. one potential byproduct of such competition is paternity confusion amongst males. Cheney and Seyfarth.. 2002. Semple et al. though at Sonso female chimpanzees are the primary perpetrators.. hence females that employ behaviours to minimise this risk will possess a selective advantage over others.. females are likely to prevent occupation by the most dominant male and increase paternity confusion amongst other males. By not advertising fertility. 1997. I propose that confusing paternity with copulation calls may therefore be one short-term vocal strategy to enlist the protective support of highranking males. Pradhan et al. in press).. O’Connell and Cowlishaw (1994) hypothesised that whilst copulation calls of baboons primarily function to incite sperm competition. 2000. 1998a. But why would Sonso females be interested in paternity confusion amongst the high-ranking males? Protection against male infanticide is of course one possibility. This is not the first time infanticide has been implicated in explaining the adaptive significance of copulation calls. these female infanticides are a direct result of increased competition amongst residents for resources. High-ranking males have been observed attempting to intervene during female-led attempted infanticidal attacks (Pusey et al.. 2006...

Specifically. compared to the more vulnerable lower ranking females. there were always more high-ranking adult females in the nearby audience. I found that when lowranking females copulated with high-ranking males and refrained from calling. By flexibly modifying call production based on who can hear them. The second aim of chapter four was to understand the other side of the problem: why chimpanzee females often refrain from producing copulation calls. Unfortunately.Chapter 7: General discussion highlighted. observations. Pers. how does female 112 . high-ranking females showed no obvious patterns. these aggression data provide some of the first evidence for males being a competed-over resource. Though the sample was low. At times. This is possibly because high-ranking females are more competitively-able and thus less likely to be as sensitive to the presence of conspecifics. females that did not employ this vocal strategy experienced substantial aggressive consequences from nearby females of higher rank (Townsend. eight instances in total. my sample of high ranking females was too small (N=2) to address this question systematically. this is the first work to invoke female-led infanticide. Whether it is their sperm or their post-natal protective services females are interested in securing is not yet clear. see chapter four). though later I address what additional studies could be conducted to further test these preliminary findings. This suggests that the “decision” to call is also mediated by the likelihood of being detected by other more competitive female conspecifics. This is certainly one aspect which could be focused on in future research. female chimpanzees may be strategically avoiding potential female initiated aggression. a very different threat. Interestingly call flexibility was only observed in the low-ranking adult females. in understanding the function of primate copulation calls.

Whilst this is probably one important aspect of copulation calling. neither call occurrence nor acoustic fine structure can be used by males to assess a female’s fertility. such ovulation concealment is predicted to create paternity confusion 113 . Similarly to chimpanzees. Chapter four also provides an additional strand of research to the question of copulation calling in primates. Pfefferle et al. Therefore. most importantly.Chapter 7: General discussion competition affect females of different ranks and do high and low ranked females employ similar vocal counter-strategies? In combination. These results provide further evidence for the hypothesis that chimpanzee copulation calls function to confuse paternity amongst males. must also be factored in and accounted for. but focused less on their production and more on what information they may additionally convey to listening individuals. Their results show that. I found that the acoustic structure of copulation calls do not indicate the female’s proximity to ovulation. 2006). I continued with the question of copulation calls as social tools. they provide no evidence regarding fertility and proximity to ovulation. these results suggest that female chimpanzees are capable of using copulation calls in flexible and seemingly strategic ways.. Recent research suggests Barbary macaques behave in similar ways. whilst calls do signal the occurrence of ejaculation. noting the timing and occurrence of calls in addition to recording them acoustically. the fact that the copulating female is part of a complex social network comprising of. Only then will we get a clearer picture of what function these vocalisations have in primates and indeed other mammals.. Most studies to date have focused primarily on these calls as signals of receptivity to nearby males (Pradhan et al. and that one potential selective pressure driving the tactical use of these calls has been competitive female social relationships. (2008b) collected over 1400 copulation events. social competitors. In chapter five.

In Barbary macaques. on the ground. 2002). in the trees) in which they were produced. but it had a significant impact when with a low-ranked male. high-ranking males do still tend to aggressively exclude lower ranking males from mating. Previously. I found no evidence for the idea that copulation calls induce male-male competition (chapter four) thus. male or female infanticide is not a relevant risk and thus paternity confusion is unlikely to enlist male protection.Chapter 7: General discussion amongst mating males. Instead. Pfefferle et al. However. I found that females would be more likely to produce them with low-ranked males if they were more exposed and if furtive mating was impossible. Although copulation calls per se may not incite competition. The reason only became evident when also taking into account the location in which these calls were produced. Brauch et al. 2007. however. 114 .. 2008a).e. why exactly females should signal the male partner was not completely clear. aided by copulation calling. Hauser. is probably the most effective strategy if females are to avoid femalefemale competition and its consequences (chapter four). 1986.. I found that females may acoustically signal the rank of the male in their copulation calls.. In chimpanzees. Location did not seem to effect a female’s production of copulation calls when with a high-ranked male. 2007. and this is thought to help other males judge the competitive ability of the mating partner. a behaviour that is known to be common in this macaque species (Heistermann et al. Watts. Yellow baboons are also known to encode such social information in their copulation calls (Semple et al. males are thought to provide a post-natal service in the form of paternal care and investment. I argued that paternity confusion amongst the high-ranking males. through employing further analyses on the acoustic structure of copulation calls and the locations (i. it became clear that males have also played a role in the evolution of this call in one way or another. 1990. usually by challenging the female (Goodall..

Perhaps due to this additional social threat.. Vocal communication is known to play a crucial role in settling agonistic contests (Anderson. 2007)..Chapter 7: General discussion 1998. one additional vocal strategy for females is to produce a copulation call. Baboons (Kitchen et al. 2007). Reby and McComb. but one that reliably indicates the rank of the partner. 1994) and it has been suggested that honest signalling in vocalisations is one mechanism through which such settling occurs (e. Van Schaik et al. Chapters four and five. Duffy et al. and conforming to social “norms”. provide the first evidence for female-specific vocal counter-strategies to social pressures. not only the social pressures females are exposed to and the kind of vocal strategies employed to deal with these effectively. such as in exposed trees. namely that of female-female competition. This is a topic I will address later. This provides a further layer to our understanding of. but also the cognitive processes that might underpin these behaviours. At the same time. 2000.g Red Deer. my results have indirectly elucidated that males. when in socially risky situations (such as mating with low-ranked males). females are sensitive to how exposed they are. 2006. matings with low-ranking males are ostensibly cryptic and females often refrain from producing copulation calls. could minimise the propensity for male aggression. However. With their copulation calls. Boesch and Boesch-Achermann. Elephant Seals (Sanvito et al. Cervus elaphus (Clutton-Brock and Albon. female chimpanzees may be honestly signalling the rank of their mating partner and it is possible such acoustic labelling. 2004)). An additional implication of these results is that.. may have 115 . Kahlenberg. 1979. and the social pressures they impose. and potentially what high-ranking males can and cannot see. in instances when furtive mating is not possible. 2003). 2004..

Unless such an integrated approach is taken. chimpanzees differ remarkably in their behaviours between communities. For example.. Very few have attempted to additionally take into account the fine acoustic structure of these calls (Semple et al. Hormonal analyses additionally showed their copulation call production was not correlated with proximity to ovulation. though recent research on the Ngogo community of chimpanzees suggests this may not be the case. 2004. confirming my results. again results in line with my findings (Gunther Pers. This highlights the very important point that. Whiten et al. or indeed other mammal species.. Whilst additional studies still need to be conducted at more contrasting chimpanzee research sites (Ngogo is roughly 250km from Budongo).Chapter 7: General discussion additionally shaped the evolution of these acoustic signals. the majority of copulation call studies have solely focused on calling frequency and the contexts eliciting their production (Pradhan et al. how they are used and what selective pressures they have evolved under. no evidence was found for copulation calls inciting male-male competition. such similar findings from a much larger sample 116 . Given the unique social and demographic structure at Budongo. To what degree similar forces determine female calling behaviour in other communities needs to be addressed.. it is very difficult to draw conclusions regarding the role these calls play in the lives of primates.. Systematic investigation into these variations has led some to suggest chimpanzees possess their own forms of culture (Whiten et al. only with in-depth analyses is it possible to begin to understand the function of animal vocalisations. as well as the instances when females do not call. it is possible that my findings regarding the function of female copulation calls are community specific. 2005). most obviously their fine acoustic structure. From a long-term data set comprising almost 6500 copulations. 2006).comm). Due to various socio-ecological factors. 2002).

it is entirely possible that differential responses are obtained because one call type is more salient than the other. Within this experimental design. 1990. Additional studies could also involve experimentally verifying the observational results presented in chapters four and five. Agonistic Screams In chapter six. Pfefferle et al.g. this is an aspect that I will come to later when discussing future directions. This is a confounding factor that is rarely tested. I moved away from how females use vocalisations as social tools.g. to try and understand how females can comprehend vocalisations and what social advantage such comprehension could give them. I found that chimpanzees can meaningfully distinguish between screams labelling differing aggression severity. Again. The tantrum control condition worked because tantrum screams matched severe screams in their acoustic parameters. when infants were not fed or carried). but consistent across at least two east-African communities. Cheney and Seyfarth. If chimpanzees were just responding to the saliency of the acoustic signal and not the informational content. in the production context.Chapter 7: General discussion size implies that my results are not necessarily Budongo-specific. To date. 2008b). Because calls can often differ in their acoustic features.. most monkey playback studies have focussed on comparing responses given only to the acoustically distinct signals of interest (e. yet were elicited under very different conditions (e. I also included an infant tantrum control condition. I predicted they should respond equally to both severe and tantrum 117 . chimpanzee victim screams are no different. In this respect.

After descending the tree WL was subjected to a coalitionary attack by alpha female NB and mid-ranking female KL.58am.59am). Because the screams played back were from offspring. adult female WL had just copulated with adult male MS. (1984) hypothesised that overt reactions to severe screams arose because it represented the greatest threat to these 118 . WL began producing victim screams and as the aggression escalated. Comprehension of agonistic severity through vocalisations is advantageous for female chimpanzees. Comprehension of victim screams is a good illustration of this more general capacity. Observations of third party. there may be other explanations. Whilst this is only one anecdote. From this. it well illustrates the point that vocal signals can serve as carriers of vital information with non-trivial fitness consequences. 1984). who was in the process of moving away. an uninvolved female KU. for example. bystander females becoming involved in female aggression are not uncommon at Budongo. KU then also began producing victim screams until the fight was dispersed by MS and high-ranking male BB. but processing the semantic content of victim screams at some level... Of course. This is the first time that acoustic intensity has been controlled for in a call comprehension playback experiment. Similar studies on Rhesus macaques found that mothers responded more to severe screams than mild screams (Gouzoules et al. became included into the aggression (11. For example on the 23rd July 2007 at 11. I suggested chimpanzees are not just responding to the peripheral acoustic features. as it allows them to keep track of changes in their surrounding social environment and to make appropriate behavioural decisions. Gouzoules et al.Chapter 7: General discussion screams. and my results showed that tantrum screams were responded to least of all conditions. subjects may be responding more to victim screams to assess whether to intervene.

As described in chapter six. at this point it is not possible to make similar claims. Unfortunately. Hence. at least in chimpanzees. KU) received the tantrum control trial. Whilst I understand this was not ideal. An additional problem with the experimental methodology relates to the control condition and the stimuli used for this. it was the only available option I had if I was to systematically address chimpanzee comprehension skill from the female’s perspective. though future studies could address this question more thoroughly. due to the arduous methodological pre-requisites required when conducting these experiments with chimpanzees. because I was interested in what information females can extract from victim screams. As the first experiment of its kind. if the focus of a playback study is again females. Consequently. but under test conditions these females refrained from travelling or resting on the ground (critical playback pre-conditions) and thus could not be completed. Firstly. I had to supplement these four females with two adult males.Chapter 7: General discussion individuals. there were some inevitable design and protocol problems that need to be rectified if future studies are to become more efficient. I was automatically restricted in my sample size. (45 non-playback days in the forest. it was still only possible to test four adult females in each condition. active steps should be taken towards pre-experimental habituation. as only a small number of adult females at Budongo are sufficiently habituated for continued following. In the test conditions. particularly in their avoidance behaviour during fights. it was only possible to test unrelated individuals over the 10 month time period. In the future. we were playing back unrelated sub-adult male 119 . 14 straight days without one experiment). Only with this measure will it be possible to test a sufficient number of female subjects. Even after lengthy attempts. at that time. Two further females (KY. I purposefully picked males that were most “female-like” in their behaviour.

The latter point is particularly true for chimpanzees. primarily because they are known to lead complex social lives (Byrne and Whiten. and indeed males. the fact remains that acoustic structures alone are insufficient to explain the observed behavioural responses Despite these methodological issues. The fact that we saw little response to tantrum screams could be because subjects generally ignore infant calls. Female chimpanzees may use these calls to judge ongoing social events. even those that exhibit fission fusion social systems. use their vocalisations as social tools in their complex daily lives. due to their fission fusion social system and the dense nature of their surrounding habitat. particularly the intensity and potential danger of ensuing agonistic interactions. Unfortunately. ensuring uniformity across all conditions. 120 . and vocalisations often represent an important communicative medium. ideally. and subsequently employ this information to make social decisions. Additional cross taxa examples Primates are a useful model for understanding how animals use vocalisations in socially calculated ways. making it impossible to get a large enough library of calls from different individuals. chapter six provided the first evidence for call comprehension of a graded signal in any wild ape species. 1988).Chapter 7: General discussion screams. With this protocol. despite their acoustic intensity. I should have used tantrum screams from the same individuals. Although this is a valid point. This chapter also showed that playbacks with wild apes are possible. such as the chimpanzee. I hope to pave the way for future experimental studies into understanding how females. Therefore. sub-adult males rarely give such vocalisations.

Mitani and Gros Louis (1998) proposed that the active modification of pant hoot structure was an important vocal strategy to enhance social bonds between males. Slocombe and Zuberbuhler (2007) investigated the potential function of chimpanzee victim screams. 121 . More recently. These results combined are particularly intriguing because they show that. 2003).. These agonistic vocalisations have long been thought to recruit aid (Goodall. Mitani and Gros Louis (1998) showed that chimpanzee males were capable of acoustically converging their pant hoots. in addition to the production of calls. these are mainly viewed from the male perspective. though. 2007). unsurprisingly. Whilst it was initially unclear as to the function of this vocally learnt behaviour.Chapter 7: General discussion So far. the acoustic structure of vocalisations can be used in socially manipulative ways. settling dominance rifts between males of differing status (Kitchen et al. not only themselves. Chimpanzees were found to exaggerate their victim screams only if there was an individual of higher rank who could potentially intervene on their behalf. males use “yelping calls” to dissuade other prospective males from attempting a take over – a strategy thought to enhance male tenure (Dunbar.. For example. As mentioned in chapter one. Baboons are another socially and vocally well-studied primate group. only a small number of studies have addressed the social functions of chimpanzee vocalisations. 1986) and Slocombe and Zuberbuhler found that such recruitment could be done in tactical ways. but also on behalf of kin (Wittig et al. Chacma baboons are now known to use vocalisations to reconcile and repair social bonds. with more detailed analyses it became apparent that such convergence occurred primarily between close allies. with an entirely different harem-based social system. 1984). Their longdistance “Wahoo” calls are also crucial. Females are also thought to use vocalisations in tactical ways producing “solicitation calls” to incite recently aggressive (and hence successful) nearby males to approach and inspect her sexual swelling. In Gelada baboons.

Chapter 7: General discussion Primates are not the only animals to employ vocalisations in socially complex ways. Dolphins (Tursiops truncates), for example, produce acoustically and individually distinctive signature whistles. These whistles are more generally thought to maintain social cohesion and facilitate the prevailing fission-fusion social system (Janik and Slater, 1998). However, similarly to chimpanzees, allied male dolphins have been found to acoustically modify their signature whistles into an “alliance signature” that serves to identify the caller as a member of the alliance (Smolker and Pepper, 1999). Exactly who these alliance signatures are aimed at and what social function they have is not clear, though a number of possibilities have been suggested. Signatures could be directed at other males. Given that male units often compete with one another (Connor et al., 1992), sharing a common vocal label might make it easier for the members of an alliance to signal that they are allied and therefore represent a formidable competitive force. Alternatively, alliance signatures could be directed towards receptive females. These whistles could indicate the males’ ability to herd cooperatively and effectively and hence decrease the likelihood females will attempt to escape (Smolker and Pepper, 1999). Whatever the explanation, it is clear that dolphins utilize their distinctive signature whistles in interesting and potentially socially adaptive ways.

Outside of the mammals, chickens (Gallus gallus) provide some of the best evidence for seemingly strategic call use. When encountering food, chickens produce stereotypical pulsatile food calls. A cockerel’s tendency to utter food calls is strongly affected by the presence and nature of the receiver (Marler et al., 1986b). Food calling was typically given in the presence of strange or familiar female hens, but completely withheld in the presence of a male. It is thought that such flexible, socially context specific calling, operates to attract receptive females to the presence of available food and hence increases mating opportunities for the caller. Interestingly, males were also found to produce food calls even when there was


Chapter 7: General discussion no food available. This study has been interpreted as evidence for tactical deception in animal communication (Hauser, 1997) suggesting that even chickens are capable of socially calculated vocal behaviour.

A common theme that seems to be lacking from these studies is how social context effects female vocal behaviour. My results from chapters 3-6 might begin to address this gap in understanding. The world through which female chimpanzees navigate is socially multifaceted, even dangerous, and one way for females to maneuver successfully within this world is through the skilled use of vocalizations. I hope that this research will stimulate further, much needed focus, on female chimpanzees and other mammal species. Only with more research effort will we truly begin to understand the complexities underlying female social behaviour and the accompanying vocalisations.

Chimpanzee communication and cognition My findings also provide a crucial insight into the cognitive mechanisms that might underlie chimpanzee behaviour. Here, in this penultimate part of the discussion, I will briefly outline what we already know regarding chimpanzee cognition and how my findings from this thesis contribute to our overall understanding of this long-debated topic.

From a number of primarily captive-based studies, we now know that chimpanzees are particularly skilled in the cognitive domain (e.g. Tomasello et al., 2003). In the social realm chimpanzees appear to be capable of differentiating self from others, passing the mark test on a number of occasions (Gallup, 1970). They are also capable of tracking lines of gaze geometrically (Tomasello et al., 1999), perspective taking (Hare et al., 2000), and even potentially understanding what others know (Hare et al., 2001).


Chapter 7: General discussion In the physical world chimpanzees exhibit equally skilled cognitive behaviour. Captive work has recently suggested that they possess causal understanding of various problems (Horner and Whiten, 2005; Hanus and Call 2008), which complements evidence from the wild showing that they are skilled tool-users. Tools may be used to extract food sources (such as nuts (Boesch and Boesch Achermann, 2000), honey and termites (Goodall, 1986)), or to attend to themselves (e.g dabbing bleeding wounds, cure stomach ailments (Goodall, 1986)). Furthermore, because of the intricacies associated with these skills, many require active learning (Lonsdorf et al., 2004). The exact mechanism underlying such learning is a particularly contentious issue. Though teaching has been suggested (Boesch, 1991a), these observations are anecdotal and no study has, as yet, produced systematic evidence for this method of skill acquisition (Slocombe, 2005). Trial-and-error learning is one alternative, as is learning skills through social observations (Whiten et al., 2004). The ability for chimpanzees to learn by copying others is thought to be a particularly useful strategy as it improves the number of novel complex behaviours they can gain over a lifetime (Whiten et al., 2005).

The results from this thesis provide further evidence for the growing point that chimpanzees are skilled cognitive animals, but primarily from the vocal realm. This is a commonly ignored avenue when assessing ape cognition (Slocombe, 2005). Firstly, my findings indicate that copulation call production is affected by the female listening audience (chapter four). To date, most audience effects have been identified concerning the mere presence or absence of conspecifics (Le Roux et al., 2007) and its influence on calling behaviour. In addition to this, female chimpanzees are also sensitive to the specific composition of the audience, from a rank-based and numerical point of view. Constantly keeping track of ever-changing audience composition, and its social implications, is no simple task (Wilson et al., 2001). It requires that females monitor their surrounding conspecifics and use this information to make vocal


my results contribute importantly to previous findings. been highlighted as potential indicators of cognitive sophistication in primates (Hauser. If this is indeed the case. Tactical behavioural strategies.Chapter 7: General discussion “decisions”. 2000). has on the behaviour of chimpanzees. 2007. This thesis proposes that females can use vocalisations in potentially tactical ways. chimpanzees may be capable of drawing inferences about their social world they cannot see from listening to and processing vocalisations. 1997. its very likely we have overlooked the role information about out-of-sight events. these results provide an extra ecological dimension – indicating one possible advantage to this skill in the wild.. These results contradict the traditional assumption that chimpanzees are purely attentive to the most salient signals in their environment and that their comprehension of vocalisations is cognitively simplistic (Owren and Rendall. Tomasello. such as deception have. 1988). Such developments provide a much-needed ecologically relevant approach for future work investigating the cognitive sophistication of our closest living relative. Whilst perspective taking has been previously shown in captivity (Hare et al. Chapter five provides an additional layer of cognitive complexity showing that females may also take into account what males can see. Instead. Findings of this sort provide some evidence against the widely held model of primate calls as hardwired response predispositions (see Zuberbuhler. 1997). Byrne and Whiten. Through focusing on the more parsimonious explanations. in press) in favour of explaining call production in terms of complex social cognitive processes. in the past. gained from vocalisations. 125 . the chimpanzee. I additionally showed that chimpanzees are capable of processing the informational content of victim screams and possibly using this to make informed future behavioural decisions.

specifically their sperm or the protective post-natal services they may provide. It is also very plausible that females may employ similar behaviours in order to maximise their access to high-ranking males. This is one hypothesis that could be directly tested with more observational data from Budongo. For example. we still have little detail regarding this social process. Evidence for such direct competition over males has been found in a number of primate and mammal species (Rosvall. including direct aggression towards soliciting males. 1998b). In this thesis.possibly as a form of competitive exclusion. females can use their copulation calls in a flexible manner. Furthermore. 2008). male primates mate guard females through a number of behavioural mechanisms. it would be important to address similar questions at other field sites. to ensure the effects we are detecting are not site-specific. maintaining close proximity to the cycling female. As previously explained in chapter five. particularly the underpinning mechanisms. these results are purely observational and it is therefore crucial to systematically verify them with a playback experiment. it would also be very interesting to isolate potentially more subtle means through which females compete over this resource. or post-coital grooming with the female (Semple. but instead part of the wider socio-ecology of the chimpanzee. My findings suggest that in response to female competition and the aggression that can accompany it. we were initially wary to attempt the first wild ape playback experiment with such a reproductively important 126 .Chapter 7: General discussion Future directions What are the next steps in understanding female competition and the accompanying vocal counter-strategies? Because female competition in chimpanzees was initially disputed. Albeit interesting. however. I showed that high-ranking females can display severe aggression to low-ranking females when advertising successful copulations . I have proposed that one resource females may compete over is males.

“Do oestrus female chimpanzees modify copulation call production in a tactical manner?” In this study I would aim to experimentally test whether females have voluntary control over their copulation calls and whether they produce the calls in a tactical fashion. If females have voluntary control over their vocal production and are sensitive to the presence of higher-ranking females. pant-grunts from a socially less relevant. reduce the chances of extinction behaviour and minimise social disturbance. as we have now managed to devise a methodologically and ethically sound protocol for conducting playback experiments. the possibility of testing the function of chimpanzee copulation calls has become increasingly more possible. low-ranking female are played back and the prediction is that this should not affect copulation call production. This is just one example of a number of different experiments that could be undertaken to test the function of chimpanzee copulation calls. Obvious additional studies include testing what 127 .Chapter 7: General discussion vocalisation as the copulation call. I predict that they should suppress the production of copulation calls when hearing pant-grunts of higherranking females. Within this pre-copulatory time window. the pant-grunt. During oestrus periods. As a control. One of the most surprising results from chapter four was that the presence of high-ranking females affects the female’s decision to give a copulation call. through playing a frequently-occurring ‘greeting’ vocalisation. female chimpanzees mate promiscuously and copulation is often preceded by male and female courtship behaviour. I will now outline a potential experiment that could rigorously test this novel result. I would employ the playback protocol outlined in chapter six to maximise realism. However. I will simulate the arrival of a high-ranking female into the party.

2008). 1998b. complement very recent findings that suggest females employ subtle yet complex behavioural strategies to avoid conflict with each other (see Emery Thompson et al. their lives periodically turned upside down by sexually-excited. environment. 2008b. one very logical prediction is that information regarding ongoing mating activity is also relevant to females (Pfefferle et al. The mere concept that they may compete with one another over resources and exhibit male-like aggressive tendencies when doing so. particularly whether they can extrapolate female identity or the rank of the mating partner. This portrait depicts females as living in a complex and volatile social world.. Together this highlights 128 . aggressive males. would have been immediately ruled out. These results. 2000. Slowly but surely. In a system where females theoretically compete for access to males.Chapter 7: General discussion information males can extract from copulation calls. peaceful individuals. Similar playbacks have been employed with other monkey species and have yielded very interesting results (Barbary Macaques: Semple. our portrait of the female chimpanzee is changing. Semple and McComb. I have attempted to help paint a new emerging picture. Such questions are likely to provide a whole new dimension to understanding the evolutionary function of primate copulation calls and indeed other socially relevant vocal signals Conclusion Only recently has the myth of the passive female chimpanzee been dispelled. Chimpanzees are no exception and I would be interested to see how they use such information to make informed social and sexual decisions. Yellow Baboons: Semple. For a long time females were considered as solitary. A further subject that is likely to receive a lot of future interest. 2001). 2008b). at times hostile. is what female conspecifics can understand from copulation calls. from the vocal domain. In this thesis.. Pfefferle et al.. capable of using and comprehending their vocalisations in strategic ways allowing safe navigation through this.

behaviours and ultimately the minds of female primates.Chapter 7: General discussion that even where females interact infrequently. I hope these findings will stimulate further research into the social lives. 129 . “female initiative” (Wrangham. 1980) can be a powerful force in the evolution of social and vocal systems. Selective impact does not necessarily correspond with social power.

H. (1994). 55: 109-132. & Alberts. (1963).G.. Sexual selection. Cheney. (1993). Cognitive aspects of travel and food location by chimpanzees (Pan troglodytes schweinfurthii) of the Budongo Forest. Science. T. (2003). Moos-Heilen. S.. & Huntingford.G.M. & Mylrea. C. Bates. M. Bergman. Altmann. R. Bisazza. Behaviour. Magnhagen. M (2000). 49: 227-267. Behavioural Ecology. Altmann. F. (1990). 4: 184–187. 227-242. F. (1993). P. B. J. C. & Reynolds. Dominance Rank.P. deWaal. Observational study of behaviour: Sampling methods. Resource Availability. W. Bahr. NI. K. J. Baker. U... Berglund. (Ed. & Zimmermann. V. American Journal of Human Biology. pp. Hodges. A. Cambridge.. (1994). & Seyfarth. Palme. 34: 233-235. A. L. Identification of wild chimpanzee hair samples from Feces by Electron-Microscopy. U. Social position and behaviour of dairy heifers in yards. R. (1993). Hierarchical classification by rank and kinship in baboons.K. R. J. Anderson. & Wrangham. 302: 1234-1236. C. R. Arcadi. McGrew. In: Chimpanzee Cultures. Variability in reproductive success viewed from a lifehistory perspective in baboons. PhD thesis. & Strum. Möhle. J. Animal Behaviour. F. and Reproductive Maturation in Female Savanna Baboons. Bercovitch.C.. Social Relationships of Female Chmpanzees: Diversity Between Captive Social Groups. König. Smuts. Folia Primatologica. NJ: Princeton University Press. Vocalizations of adult gelada baboons (Theropithecus gelada): acoustic structure and behavioral context. Behavioural Ecology and Sociobiology.C. Female– female competition over reproduction. C. . E. (1999). (2005).. Andrews University. J K & Heistermann.L. B. D. 15(3): 401-409. C. St. Beilharz. Primates... H. Infanticide in chimpanzees: Review of cases and a new within-group observation from the Kanyawara study group in Kibale National Park.).J. Beehner. General and Comparative Endocrinology. 130 . A. 117: 427-438. S.W.. Primates. Massachusetts: Harvard University Press in Cooperation with the Chicago Academy of Sciences. R. (1974).. Bakuneeta.. 11:522.. 33(5): 313-318. and Heltne. Inagaki. (2003). R. W. Princeton. Comparative aspects of the metabolism and excretion of cortisol in three individual nonhuman primates. Uganda. 40: 337-351. B.. by Wrangham.References and Appendix References Aich.

R. CA. C. C.. Female sexual behavior and sexual swelling size as potential cues for males to discern the female fertile phase in free-ranging Barbary macaques (Macaca sylvanus) of Gibraltar. by A. T. Nature. D. (1972). American Journal of Primatology. 131 . N. J. 41: 530-532. (2000). M.. R. d. W. 117: 220-242. Hormones and Behaviour 52: 375–383. Boesch. Boesch. M... apes. Teaching among wild chimpanzees.. A. (1991b). Menlo park. Oxford: Clarendon Press. and Heistermann. & Whiten. (1996). A. K. (1988). In: Coalitions and alliances in humans and other animals (Ed. & McCown.D. Waal). Range Use of the Forest Chimpanzees of Kibale . Gabon. (1993). & Boesch. C. C. Tagg. Nature. R. Chapais. Cheney. Möhle..A. Chapman. T.& Boesch-Achermann. The role of alliances in social inheritance of rank among female primates. R. L. pp. Harcourt & F. The Nut-Cracking Behavior and Its Nutritional Importance in Wild Chimpanzees in the Tai National-Park. J. New York: Oxford University Press. and Robbins. Byrne. (1990). D. D.Arandjelovic. Agonistic behaviour. K. Ivory-Coast. Chicago: University of Chicago Press.. (2008). (2007). U. dominance. The chimpanzees of the Tai Forest. D. Pfefferle.References and Appendix Boesch. Behaviour.. H. (1984). Cannibalism among wild chimpanzees. & Slagsvold. Hodges. Breiehagen. Fatal Chimpanzee Attack in Loango National Park. (1988). (1979). Animal Behaviour. H. M. (1991a). Bygott. Fishcer. 405-428. 29-59. Vigalent. (Ed.B. C. & Seyfarth. Boesch. by Hamburg. International Journal of Primatology. 31(4): 263-273. Head.. International Journal of Primatology. J.Implications for the Understanding of Chimpanzee Social-Organization. 238: 410 Bygott. Animal Behavior: the question of culture.. pp. The effects of leopard predation on grouping patterns in forest chimpanzees. Benjamin/Cummings Publishing Company. Animal Behaviour. J. B.). 28:1025–1034 Brauch. E. 5(4): 323-323. & Wrangham. 36: 604–605. C... Oxford University Press. Boesch. and humans. In: The great apes. Boesch. L. Machiavellian intelligence : social expertise and the evolution of intellect in monkeys. Male polyterritoriality and female–female aggression in pied flycatchers (Ficedula hypoleuca). How monkeys see the world.M.. C. (1992). 379: 207-208. and social structure in wild chimpanzees of the Gombe National Park.

Timing and probability of ovulation in relation to sex skin swelling in wild West African chimpanzees. & Silk.R & LeBoeuf. & Seyfarth.A.... & Boesch. Heistermann. 28: 129-141 Dewsbury. Hiraiwa-Hasegawa.A.C. M. 89(3): 987-990. London: Cape. University of Chicago Press. 66:551-560. S. The responses of female baboons (Papio cynocephalus ursinus) to anomalous social interactions: Evidence for causal reasoning? Journal of Comparative Psychology. & Boesch. Animal Behaviour.References and Appendix Cheney. B. Green. D. S. (1992). Primate ecology and social organization. T.F. J. and Richards A. B. Cheney. Cox.. D. 119: 601–610 Deschner. Perineal swelling during gestation and maternal competence in chimpanzees. R. Journal of Zoology. R. lek breeding and mate choice in fallow deer. 183: 1-39. (2007). Smolker.. (1991). F. 111: 317-335. (2004). Hodges. Creel. Hormones and Behaviour. Heistermann. 43:1035–1044 Clutton-Brock.F (1999). & Albon. P.. Ejaculate cost and male choice. Clutton-Brock. (2003). D. The evolution of parental care. 45: 204-215. Female sexual swelling size. 66: 115-125. J. Female incitation of male competition: a mechanism in sexual selection. timing of ovulation and male behaviour in wild West African chimpanzees. (1995). Behavioural Ecology and Sociobiology.. M.). 109: 134-141. Clutton-Brock. (1979). R. Journal of Medical Primatology. Animal Behaviour. For whom does the female dove coo? A case for the role of vocal selfstimulation. K. American Naturlist. C. H. Context-specific calls in wild chimpanzees. C. 86(2): 255-264. (1992). and Boesch. (Pan troglodytes verus): analysis of barks. Crockford. H. (1988).L.. T. M. Hodges. D. Seyfarth. T. Connor. Two levels of alliance formation among male bottlenose dolphins (Tursiops sp.B. C.. Passing the buck: resource defence. D. M. R. Press. Dominance. (1982). (1982). K. Chimpanzee politics power and sex among apes. Dahl. C. & Albon. 50: 145-170.F. (2003).L. Chicago. Journal of Mammology. Baboon Metaphysics: The Evolution of a Social Mind. Behaviour. Proceedings of the Royal Society of London B. F. C. (2005). T.M. Deschner.. H. Princeton: Princeton Univ. T. Cheng. (Pan troglodytes verus). Clutton-Brock.. (1977). D. and glucocorticoid levels in social carnivores. The roaring of red deer and the evolution of honest advertisement.. 23: 281-296. (1977). H.M. S. T. aggression.J. H. De Waal. Animal Behaviour. M. & Harvey. 132 . American Naturlist.

The social brain: Mind. 27: 1601-1617 Emery Thompson. (2006).E. Journal of Ecology. M. Comparison of sex differences in gregariousness in fission-fusion species: reducing bias by standardizing for party size. J). group size and language in humans. In: Primates of Western Uganda (Ed. Ethology and Sociobiology. Dunbar. Journal of Zoology of London. (1993). Core area quality is associated with variance in reproductive success in chimpanzees at Kanyawara.. 16: 257-333 Emery Thompson. (1995). I. M. 32: 163-181. Eggling. and Silk. 16: 681-694. 29: 816-822 133 . East. and Wrangham.L. W. Dunbar. Uganda. H.E. Duffy. Emery Thompson. L.M.J. M. W. New York: Kluwer Academic. 694-735. Stumpf. R. M.. ranging.. R. 28 (2): 389-406 Doran.. Princeton University Press Dunbar. 17(15): R586-R587 Dunbar. (2007). by Newton-Fisher. Notman. (2007). M. Kibale National Park. and Boesch. (1984). A. Can the patterns of sexual swelling cycles in female Tai chimpanzees be explained by the cost-of-sexual-attraction hypothesis? International Journal of Primatology. R. H. & Sharman. 219: 690–697. R. I. Annual Review of Anthropology. Kahlenberg. International Journal of Primatology. Animal Behaviour. K. N.W. V. Dominance and reproductive success among nonhuman animals: a crossspecies comparison. (1997). Co-evolution of neocortical size. M. 34:20-87. and Turk. 73: 501-512. (1983). J. R. (2007). Wrangham. and Pusey. Behavioural and Brain Sciences. Reynolds.References and Appendix Deschner. Hofer. S. Influence of seasonality on activity patterns. I. Newton-Fisher. C. International Journal of Primatology.M. R. (1947). and grouping patterns in Tai chimpanzees.. (2003). N. Behavioural Ecology and Sociobiology. language. I. 13(2): 157-159. feeding behavior. Female reproductive strategies and competition in apes. (1989). and Paterson... T. V. Functions of birth dens in spotted hyaenas (Crocuta crocuta). Reproductive Decisions: An Economic Analysis of Gelada Baboon Social Strategies.. A. 18(2): 183-206. R. Female Competition for Access to Males Affects Birth-Rate in Baboons. M. Probable community transfer of parous adult female chimpanzees in the Budongo Forest. Ellis. and society in evolutionary perspective. Gilby. Emery Thompson. Uganda. D. and Reynolds.. M. Current Biology. M . Male chimpanzees exchange political support for mating oppurtunities. (in press). & Wrangham.C. Observations on the ecology of the Budongo rain forest. R. International Journal of Primatology.E (2008).

Vocal communication: a vehicle for the study of social relationships.. & Marler. Primates. and grouping of wild pygmy chimpanzee females at Wamba. Gage M. R. 51:243-247. Zaire. Science. Tanzania 1967-1978. 28: 259-282. (1970). (1964). (2000). Goodall. Furuichi. Zeitschrift Fur Tierpsychologie. Baboon responses to graded bark variants. M. Rinehart and Winston. K. Chimpanzees: self-recognition. and Cook P. S. Gouzoules. S. H. 30: 1138–1151. R. by Jay. L. (1986). 61: 925-931. PhD thesis. Sexual swelling.References and Appendix Fawcett. L. In: The Cayo Santiago macaques (Ed. J.. (1994). (1968). The Chimpanzees of Gombe: Patterns of Behaviour. Sperm size or number? Effects of nutritional stress upon eupyrene and apyrene sperm production strategies in the moth Plodia interpunctella (Lepidoptera: Pyralidae).). Fawcett. J. J. 111-129. Animal Behaviour.A. Frame. J. (2001). Animal Behaviour.G. H. Folia Primatoligica. W. Frame. Gallup. Gouzoules. T.G. New York: Holt. 182-193. M. (1984).. UK. P. Edinburgh University. Infant Killings and Cannibalism in Free-Living Chimpanzees. Goodall. & Van Lawick. 8: 594-599. (1977).. pp. P. Goodall. Tool-using and aimed throwing in a community of free-living chimpanzees. Animal Behaviour. 134 . R. Fischer. Cheney. 50: 225-249. G.. Female relationships and food availability in a forest community of chimpanzees. 201: 1264-66. Metz.J. A.G. & Seyfarth. In: Primate studies in adaptation and variability (Ed. L. Gouzoules. H. J. & Kessler.).. & Marler. 23/3: 309-318. G. Death of a wild chimpanzee community member: possible outcome of intense sexual competition.. and Fedigan. Functional Ecology. (1987). (2000). Gouzoules. Behavioural dominance and reproductive success in female Japanese monkeys (Macaca fuscata). receptivity. G. Malcolm. H. Gouzoules. American Journal of Primatology. & Muhumuza. (1982).J.. D. 32. P. M. 167: 86–87. Rhesus monkey (Macaca mulatta) screams: Representational signalling in the recruitment of agonistic aid. Gouzoules.. (1979). Cambridge: Harvard University Press. (1986). Social organization of African wild dogs (Lycaon pictus) on the Serengeti Plains. A preliminary report on expressive movements and communication in the Gombe Stream chimpanzees. K. H.. J. S. Nature. by Rawlings. Albany: SUNY Press. Goodall.

Bristol. A.. 112-143. Hare. Cambridge. Mass. & Arrowood. (2008). Call. C. (1987).. L. M. Cambridge: Cambridge University Press.C. Tanzanian Journal of Ethology. R. Journal of Zoology. (1990).). pp. (2001). Lawrence Erlbaum Associates.. Gust D. Animal Behaviour. (1992). Dominance and fertility among female primates. Chimpanzees using stones to crack open oil palm nuts in Liberia. Hamilton. (Ed. (1991). In: Machiavellian Intelligence II. Griffin. Agnetta. (2005). Griffin. (1997). A.. (2007). Evolution and Communication. Call. Cognitive Ethology. (1998). (1983). M.. Primates. 200: 1405-1409.: MIT Press Hauser.I.D. Hanus. (Ed. Hauser.. Current Biology. Whiten. gibbons (Hylobates hoolock). & Byrne. D. C. P. (2000). 2:189–215 Greengrass E. M. Hannah. Animal Behaviour. M. W. J. 39: 596-597.. Hare. 1: 75-85. Minding the behavior of deception. 18(9): R370-R372 Harcourt.D. 254: 93-96. A. B. H.. Do chimpanzees know what conspecifics know? Animal Behaviour. 73(6): 10591065 135 . Opportunistic and restrictive matings among wild chimpanzees in the Mahale mountains.. T. copulation calls. 33: 151-162. Copulatory vocalizations of chacma baboons (Papio ursinus).A. UK).. 61: 139-151. 59: 771-785. Nishida. Primates. Sociability and dominance among female chimpanzees at Gombe (University of Bristol. Chimpanzees know what conspecifics do and do not see.A.D. W. Hasegawa.J. St. & McGrew. Science. B. by. Hauser.. Andre E. H. M. & Turner. by Ristau. & Tomasello. D. M. B. 28: 31. T. Rhesus monkey (Macaca mulatta ) copulation calls: Honest signals for female choice? Proceedings of the Royal Society of London B. New records of within-group infanticide and cannibalism in wild chimpanzees. M. & Hiraiwa-Hasegawa. Call. (1987). (1993). J. (1996). The evolution of communication. M. Donaghey B. A. Takasaki. (1978). Hillsdale. Hauser. & Tomasello.). Do Chimpanzee copulatory calls incite male-male Competition? Animal Behaviour.W. and humans. When males call. Chimpanzees infer the location of a reward based on the effect of its weight. C. Hauser. NJ Hamai. females listen: sex differences in responsiveness to rhesus monkey (Macca mulatta). In: The Minds of Other Animals Essays in Honor of Donald R. M.D. Estrus vocalizations in two primate species (Cercocebus torquatusatys and Macaca nemestrina): evidence for an effect of intrasexual competition. 213: 471-487. J.References and Appendix Gouzoules H..

S.References and Appendix Hausfater. K. American Journal of Primatology. (1983). The Social Function of Intellect. Application of urinary and fecal steroid measurements for monitoring ovarian function and pregnancy in the bonobo (Pan paniscus) and evaluation of perineal swelling patterns in relation to endocrine events. Horner. 8(3): 164-181. B. M. by Bateson. S. M. Heistermann. Hrdy. Group fusions and minimum groups sizes in vervet monkeys (Cercopithecus aethiops). Maple.). Germany. Hiraiwa-Hasegawa. Academic Publ.L.. PhD Thesis. T. F. Heistermann. New York . Hrdy.. G. R. Cheney. (1994). Möhle. pp 303-317. D. Biology of Reproduction. 69:1–15. Möhle. & Whiten. Animal Behaviour. Press. & Harwood. U. Massachusetts. 137-154.K. 51: 233–234. S.. H. (1996). R. & Seyfarth. New York: Academic Press. Animal Cognition. J.. Hoff.D. Infanticide: Comparative and Evolutionary Perspectives. C. Vervaecke. (2005). Behavioral biology and the double standard. S. S. Brauch. American Scientist. (1996). pp. & Hodges. American Journal of Primatology.K. Cambridge University Press: Isbell. and Hinde. (2007). In: Infanticide and Parental Care. Hodges. The Woman That Never Evolved. UK. G. van Elsacker. 25: 57–65.A.. Herbinger..S. Infanticide in non-human primates: sexual selection and local resource competition. and Hrdy.K.. & Williams. N. L. U. A. Henzi. Causal knowledge and imitation/emulation switching in chimpanzees (Pan troglodytes) and children (Homo sapiens). 65: 4049. In: Growing Points in Ethology.. Harvard Univ. 55: 844-853. In: The Social Behavior of Female Vertebrates (Ed. University of Leipzig.P. B. Nadler. L. Pfefferle.). J. London.. (1985).. 38:72-85. Aldine. (1984). (2004). Humphrey.). Folia Primatoligica. K. D. M. Cambridge.P. vom Saal. (1977). (Ed. V. (Ed. Female ovarian cycle phase affects the timing of male sexual activity in free-ranging Barbary macaques (Macaca sylvanus) of Gibraltar. 3-17. Infanticide as a primate reproductive strategy. 136 . A.B (1981). Copulation calls and paternity in chacma baboons. P. S. T. Hrdy. by Wasser. (1991). G. I.B.. pp. R.. Inter-group aggression in wild West African chimpanzees (Pan troglodytes verus): Mechanisms and functions. L. (1976). M. Control role of an adult male in a captive group of lowland gorillas. by Parmigiani.. Hasegawa. Dittami and J..

S. Lynn E. Hatchwell B. Kahlenberg.M. A. Proceedings of the Royal Society of London B.. E. (1971). Emery Thompson. Nature. Animal Behaviour. Janson. P.with Special Reference to the Newly Observed Case in the Mahale Mountains.. (2004). A. 23: 759-783. Langmore N. Koenig. L. D.References and Appendix Janik. Hartley I.R. PhD thesis. J. (in press b). and Wrangham. V. H. Female competition over core areas among Kanyawara chimpanzees. Uganda. (2002). Primate societies : group techniques of ecological adaptation. S. 428: 715-716. & van Schaik. & Slater. M. Eberly. Manser. M. (Papio cynocephalus ursinus). (1996). Female song attracts males in the alpine accentor (Prunella collaris). Kahlenberg. R.E.. Loud calls as an indicator of dominance in male baboons. & Cheney. C. H.N. Context-specific use suggests that bottlenose dolphin signature whistles are cohesion calls. 105: 165–186. Kitchen. 197218. (1998).M... Immigration costs for female chimpanzees and male protection as an immigrant counter-strategy to intrasexual aggression. Ill. M. Animal Behaviour. Cherry. Behaviour. Fischer. Lonsdorf. M. S. Seyfarth. Competition for ressources and its behavioral consequences among female primates. (2003).. R. Kummer.. Kibale National Park. Cheney.. Harvard University. International Journal of Primatology Kahlenberg.W. M.. Arlington Heights. Recognizing the many faces of primate food competition: methods. A. Kawanaka. Behaviour. (1981). & Seyfarth. (1988). Vol. & Pusey. the yellow mongoose..M. D.. Animal Behaviour. 53: 374-384 Kitchen. Female-female competition and male sexual coercion in wild Kanyawara Chimpanzees. (2004). & Wrangham R. USA. Behavioural Ecology & Sociobiology. 56: 829-838. C. Davies N. (2006). International Journal of Primatology. M.B.. D. Infanticide and Cannabalism in Chimpanzees . Sex differences in learning in chimpanzees. (Cynictis penicillata). Factors mediating inter-group encounters in savannah baboons (Papio cynocephalus ursinus).: AHM Pub. 137 . Emery Thompson. P. A.M.W.. 1: 69-99.L. V.J.. African Study of Monograms. K. 141. 75(3):943949. Muller. R. J.. A.M. (in press a). V. M I. Le Roux. D. The audience effect in a facultatively social mammal. M B (2008). 263:141–146.

16: 106-113. Hamai. Variation among mammalian alarm call systems and the problem of meaning in animal signals. Marler. J.. & Hinde. A. by Bateson. A. P.J. K. (2007). (1989).. R. pp. J. (2005). 93: 177-197. Behavioural Ecology and Sociobiology..) Cambridge: Cambridge Univ. (2005)... P. Mitani. Coevolution of vocal communication and sociality in primates. C. (1986b). Bloomington: Indiana University Press.. P. & Evans. Vocal communication in the domestic chicken: II. McComb. J. 61:661-668.. 1(4): 381-385. & Semple.). Social organization.. (1999)..W... S. 3: 501-508. (1976).D. Signaling behavior of apes with special reference to vocalization. H. Journal of Reproductive Fertility...A. (Pan troglodytes schweinfurthii). T. Matsumoto-Oda. Variations in scent marking and ultrasonic vocalizations by Long-Evans rats across the estrous cycle. Hunt. R. Behavioural Ecology..A. Hayaki. In: Growing points in ethology. & Pickert. R. C. Estrus cycle asynchrony in wild female chimpanzees. (1970). T. 51:783-786. Marson J. 965-1033. Vocalizations of East African monkeys. Primate copulation calls and postcopulatory female choice. Takasaki. 13: 81-91. (1974). (1993). Marler. R. E. I. Influence of ejaculation frequency on semen characteristics in chimpanzees (Pan troglodytes). Is a sender sensitive to the presence and nature of a receiver? Animal Behaviour.. Mitani. In: How animals communicate (Ed. Chorusing and call convergence in chimpanzees: tests of three hypotheses. Behaviour. Journal of Human Evolution. K. Dufty.A.. & Roney. Kasuya.. Cooper R. Physiology and Behaviour. & Takahata. (1998). Biology Letters. Jouannet P.. P. N. & Bar¢eld. (1977).R. D. 138 . J. M. Ethology. A. Folia Primatologica. Hosaka. Gervais D. S. communication and graded signals: The chimpanzee and the gorilla. H. Female choice in the opportunistic mating of wild chimpanzees (Pan troglodytes Scweinfurthii) at Mahale. P. W. 46:258-266. Tool use by wild chimpanzees in feeding upon driver ants. Kawanaka. Matsumoto-Oda. White. Press. Meuris S. Y. Red colobus. 135: 1041-1064. Marler. & Tenaza. K. R. 85: 43-50 Matochik. Maestripieri. by Sebeok. J. Marler. 34: 194-198. (1992).References and Appendix Macedonia. McGrew. Behavioural Ecology and Sociobiology... (Ed. & Gros-Louis.. J. M.

S. Newton-Fisher. and Fischer. (1982)..028. Agonistic relations among Kanyawara chimpanzees. Newton-Fisher.2006.M. M. and Pusey. (2006).E.W. Sommer C. Muller M. and Wrangham.:1016/j. African Journal of Ecology. (2006).F. M. 274:1009-1014 Mundry. and Kramer D. 28:23– 37. 35: 275-331). Male coercion and the costs of promiscuous mating for female chimpanzees. Discriminant function analysis with nonindependent data: consequences and an alternative Animal Behaviour. N. Is sperm cheap? Limited male fertility and female choice in the lemon tetra (Pisces. Behavioural Ecology. (1977). Pusey A.W. L. K. Science. The diet of chimpanzees in the Budongo Forest Reserve.E. 139 . D. (2007). J. Wroblewski E. E. R. On the occurrence and significance of motivational-structural rules in some bird and mammal sounds. & Wrangham. (1999a). N. International Journal of Primatology. Morton..12. 216: 753–755. R. Infant killers of Budongo. Proceedings of the Royal Society of London B. 70: 1079-1086. Muller M. 16: 2234–2238. A. 111: 855-869.N. New case of intragroup infanticide in the chimpanzees of Gombe National Park. & Watts. C. PhD Thesis. 56: 256-259. 70: 167-169. & Mitani. (2005). (2002). (1998). Muller. by Boesch . Conflict and cooperation in wild chimpanzees. Mundry R.M.References and Appendix Mitani. Foraging strategies as a function of season and rank among wild female chimpanzees.E. L. Correlates of territorial boundary patrol behaviour in wild chimpanzees.. P. C. 37: 344-354. Murray. M. (2007). Emery Thompson.. Use of Statistical Programs for Nonparametric Tests of Small Samples Often Leads to Incorrect P Values: Examples From Animal Behaviour. Nakatsura. Uganda. R.. Current Biology. Hohmann. G. (2005). Eberly.L.C. J. (2007). E.. Characidae).. J. N. M.E. Kahlenberg S. and Marchant. C.. Newton-Fisher.anbehav. In: Behavioural Diversity in Chimpanzees and Bonobos. UK. Animal Behaviour.. Muller. University of Cambridge. Folia Primatoligica. Tactical behaviour and decision making in wild chimpanzees. N. E.. American Naturalist.) Cambridge University Press. Animal Behaviour. 17(6):10201028 Murray C. (Ed. Male chimpanzees prefer mating with old females. (1997). In Advances in the Study of Behaviour. Emery-Thompson.N. (1999b).

(Ed. M. van Hooff J. 23: 66-74. (2003)..B.S. & Rendall. Animal Behaviour. by Heltne. Nyundo. Chimpanzees and Bonobos: Cooperative Relationships among Males. Journal of Human Evolution. In: Primate Societies. R. 68-89. by Harcourt. N..References and Appendix Newton-Fisher.M.. Journal of Human Evolution. E. K. T. Westbury Academic & Scientific Publishing. (1979). Cheney.. 27: 1589 1599 Nikitopoulos E. Predatory behaviour among wild chimpanzees of the Mahale Mountains. (1989). (Ed. (1987). R.. N. Ranging patterns of male chimpanzees in the Budongo Forest. and Struhsaker. pp. Within-group cannibalism by adult male chimpanzees. 11: 73-99.L. Newton-Fisher. K. International Journal of Primatology. (2004). H. 20:1-20. 140 .. Uganda. L. Uganda: range structure and individual differences. Nishida T. International Journal of Primatology.T). African Journal of Ecology. Female coalitions against male aggression in wild chimpanzees of the Budongo forest. S. Newton-Fisher. 2: 357-370. Seyfarth. Natural history of a tool-using behavior by wild chimpanzees in feeding upon wood-boring ants.A. The home range of the Sonso community of chimpanzees from the Budongo Forest. and Hiraiwa-Hasegawa. (1973). Nishida . Primates. In: New Perspectives in Primate Evolution and Behaviour (Ed. The ant-gathering behaviour by the use of tools among wild chimpanzees of the Mahali Mountains... In: Understanding Chimpanzees. M. (1982). & Marquardt. 25:659–677 Nishida. 9: 167–224 Nishida. Kawanaka. (2005). T. T. Nishida. T. Harvard University Press Cambridge. University of Chicago Press. Wrangham.). Primates.. Contextual variation in chimpanzee pant hoots and its implications for referential communication. & Sherwood. Sterck E. (1968). Primates. 41: 150-156.). R. D. P. C. & Hiraiwa. Social interactions between resident and immigrant female chimpanzees. 287-308. (2005). 26: 274-284. Norikoshi. E. N. Arnhem E. One observed case of cannibalism among wild chimpanzees of the Mahale Mountains. B. 70: 177-190. B. pp. Uehara. D. The social group of wild chimpanzees in the Mahale Mountains. Influence of female copulation calls on male sexual behavior in captive Macaca fascicularis. Primates. Nishida. (1985). T. E.W. T. (2002b). T.. by Smuts. Notman. Nishida. (1982). Mass.G.

& Masataka. C. Female-female competition for male 'friends' in wild chacma baboons. M. sperm competition and mate choice: the function of copulation calls in female baboons. L. Mitani. & Cowlishaw G. N. General gregariousness and specific social preferences among wild chimpanzees.2007. (2001). Function of copulatory vocalizations in mate choice by female Japanese macaques (Macaca fuscata). (2008a). Infanticide and the evolution of male-female bonds in animals. Pfefferle. .. In: Infanticide by Males and Its Implications (Ed. (1995). A.12.. & Rendall. R. New York: Plenum Press. Cheney. J. 77:353-35836. (Ed. Seyfarth. Packer C. R. Hodges J. Folia Primatologica. M. (1997). Pepper. An affect-conditioning model of nonhuman primate vocal signalling. 20(5): 613-631. M.anbehav. M. P. Eberly L. Male Barbary macaques eavesdrop on mating outcome: a playback study. C. 275: 571578 Pfefferle. by van Schaik.1016/j. D..J. Experimental and observational evidence. Owren. Cambridge University Press. 54: 599-614. (2001)...).E. 299-346. 293: 690-693 Palombit. Fischer. Mixed-Effects Models in S and S-PLUS. and Janson. & Cheney. Heistermann. 239-268. (2000). The adaptive value of "friendships" to female baboons. R.. J. New York.A. (1997). (1994). Egalitarianism in female African lions Science.R. (Papio cynocephalus ursinus).003 Pinheiro J. Animal Behaviour. Cambridge Palombit. 61: 1159-1171. (1999). Hodges.. Springer. C. K. High Frequency of Postcoital Penis Cleaning in Budongo Chimpanzees. Oda. D. K.. and Bates. J. R. D.M. D.. J. & Watts. Brauch..C. Folia Primatologica. International Journal of Primatology..P. pp. (2006). Animal Behaviour.. W. R. Animal Behaviour. M. R. D. Animal Behaviour. In: Perspectives in ethology. by Owings.. (2000).E. D. Female Barbary macaque (Macaca sylvanus) copulation calls do not reveal the fertile phase but influence mating outcome. & Seyfarth. H. & Lee P.References and Appendix O'Connell S. J.K. doi: 10. 48: 687-694. Infanticide avoidance. Proceedings of the Royal Society of London B..). D. D. pp. 64:132–139 O’Hara S. L. (2008b).C. & Fischer. Palombit. 141 . Pusey A. A.M. Heistermann.

C. In: Primate Behaviour: Field studies of monkeys and apes (Ed. (in press). & Seyfarth. (1980). 145: 251-275 Preston. 65: 519–530. 28: 543-582. D.P. Anatomical constraints generate honesty: acoustic cues to age and weight in the roars of red deer stags.. 22:385-392. Chimpanzees of the Budongo Forest... Stevenson. International Journal of Primatology. I. by DeVore.. B. Murray. A. M. A. (1965). Fitch WT & Clutton-Brock TH (2005). Pusey. Reynolds... J Zool. C. Reynolds. and van Schaik. A. Infanticide-driven intersexual conflict over matings in primates and its effects on social organisation. (1994). 228: 695-699.. C. (2003).E. Pusey. Nature. Animal Behaviour. van Schaik. Bird Conservation International.B. The social contexts of some very low frequency calls of African elephants. Cheney. K. (1997). Reby.L. Zaire. V. Pusey. G. Chimpanzees in the Budongo forest. and McComb.. A. Behavioural Ecology and Sociobiology. (1988).R. Non-offspring nursing in social carnivores . (2006). (2008). Pemberton. 7:317–329 Poole.minimizing the costs.. Wroblewski. Science. & Goodall. Inbreeding avoidance in chimpanzees. Rinehart and Winston.E.L. Red deer stags use formants as assessment cues during intrasexual agonistic interactions. Journal of Comparative Psychology. C. Darwin C. 59: 333-343.. I. J. D.. pp. 114: 3646. Dominant rams lose out by sperm depletion. (1997).M. K. Williams.M. Cargnelutti B. C. E. W. 5: 362-374. G. J. J. 142 .H. 277: 828-831. 1962-1992. Wilson. Rendall. Payne.. Behaviour. J. & Wilson K. Severe aggression among female chimpanzees at Gombe National Park. (2000). A. D. Tanzania.M. 272: 941-947.. Behavioural Ecology. D.. Wallauer. & Maestripieri. W. McComb K.. V. Pusey.).. (2001).T. Reby D. A.J. The influence of dominance rank on the reproductive success of female chimpanzees. Pradhan. Behavioural Ecology and Sociobiology. E. Pradhan. R. Moss. Jr. Proximate factors mediating ‘contact' calls in adult female and their infants. 409: 681-682. (1992). J.. The evolution of female copulation calls in primates: a review and a new model. Shifting cultivation along the Trans-African Highway and its impact on the understorey bird community in the Ituri Forest. New York: Holt. Proceedings of the Royal Society of London B. Engelhard. & Packer.E. Animal Behaviour.References and Appendix Plumptre. 368-424. Langbauer.R.

E.. The function of Barbary macaque copulation calls. K. (1998b). Vocal communication by the rhesus monkey (Macaca mulatta). Cheney. K. T. Animal Behaviour. 119(1): 67-77 Slocombe. Proceedings of the Royal Society of London B. (2005). Sanvito.. & Cheney. & Altmann. C. and Zuberbühler. Animal Behaviour. R. E. 56: 43-56. PhD thesis. 15 (19): 1779-1784. Animal Behaviour. Rood. 61: 1023-1028. Functionally referential communication in a chimpanzee. Journal of Comparative Psychology. Annual Review of Psychology. V. G. M. Proceedings of the Royal Society of London B. Seyfarth. In: Reproductive Biology of the Great Apes (Ed. Oxford University Press. D..References and Appendix Reynolds. PhD thesis. K. Animal Behaviour. G. S. Cambridge University Press. Perception of female reproductive state from vocal cues in a mammal species. Signalers and receivers in animal communication. et al. The Chimpanzees of the Budongo Forest: Ecology. Mating relationships and breeding suppression in the dwarf mongoose. 267: 707-712 Semple. (1980). Slocombe. Female copulation calls in primates. Current Biology. S. R. Chicago. R. E. S.. M. (2005a). by Graham. Individuality and male discrimination of female copulation calls in the yellow baboon. A. K. 28(4): 1070-1094. Informational content of female copulation calls in yellow baboons. UK. K. (1980). Galimberti. Alberts. L. 54. 138: 279-294. (2000). Seyfarth. The Serengeti lion: a study of predator–prey relations. Behaviour and Conservation. Vervet Monkey Alarm Calls: Semantic Communication in a Free-Ranging Primate. Vocal communication in chimpanzees (Pan troglodytes). UK. Sexual selection on aggressiveness in females: evidence from an experimental test with tree swallows. (1998a). 73(2): 287-299 Schaller. Semple. (1972). V. 75:1603-1610 Rowell.H. Agonistic screams in wild chimpanzees vary as a function of social role. McComb. Animal Behaviour. (2003). American Journal of Primatology. Rosvall. S. Proceedings of the Zoological Society of London. P.. University of Chicago Press Semple. and McComb. L.B. K. R. 265: 287-291. D. 319-341. & Hinde. (2005).E. (2002). K. Miller. S. (2008). K. University of Sussex. (2001).). J. E. (1962). Semple. 28: 143-150. and Zuberbühler. Vocal signalling of male southern elephant seals is honest but imprecise. Short. (1981). Sexual selection in man and the great apes. (2005b). pp. St Andrews University. J. S. S.. 143 . 145-173. Slocombe. (2007). Semple.

In: The Behavior of Communicating. Tanzania. J. 44: 161-170.. and Zuberbühler. (1979a). The evolution of female social relationships in nonhuman primates. 72: 989-999 Slocombe. (1987).104 (43). Hare. Côte d'Ivoire.). Sterck. H..Newly observed intragroup infanticide and cannibalism in Mahale National Park. pp. J. 224.(1971). Koman J. 144 . 79: 30-48. Takahata. Behavioural Ecology and Sociobiology. Chimpanzees. M. (1979b). 58: 769-777. 57: 511-524. Watts. K. by Smith. Smolker. K. P. Guinea. follow gaze direction geometrically. Sugiyama. J Anthropological Society of Nippon. PNAS. E. (1977). RM & Boesch. Sugiyama Y. E.. & Pepper. 17228-17233 Smith.). Whistle convergence among allied male bottlenose dolphins (Tursiops sp. Y. B. A preliminary list of chimpanzees' alimentation at Bossou. Tool-using and making behavior in wild chimpanzees at Bossou. Ethology. Chimpanzees modify recruitment screams as a function of audience composition. An Ethological Approach (Ed. D. and Zuberbühler. Sugiyama Y. MA: Harvard University Press.. B. Cambridge. 42(3): 179-198. Stumpf.. T. W. Animal Behaviour. Contextual sources of information. Primates. R. W. J. (2006) Food-associated calls in chimpanzees: Responses to food types or food preferences? Animal Behaviour. A..J. (2007).. Sterck. & Koman. Zeitschrift Fur Tierpsychologie. Suzuki.References and Appendix Slocombe. Tomasello. Behavioural Ecology and Sociobiology. 20: 513-524. (1997). & Leland. 69: 89-132. 28: 133-147. Carnivority and cannibalism observed among forest-living chimpanzees. P. (1999). W. 41(5): 291-309. Primates. Y. Guinea. Koman J. Primates.. (Pan troglodytes). Determinants of female dispersal in Thomas langurs. Social structure and dynamics of wild chimpanzees at Bossou. L. K. (1999). & vanSchaik. (1997). Does promiscuous mating preclude female choice? Female sexual strategies and mate preference in chimpanzees of the Taï Forest. Folia Primatoligica. Struhsaker. American Journal of Primatology. C. E. M. & Agnetta. C (2005). 105(7): 595-617. H. E. Adult Male chimpanzees kill and eat a male newborn infant . (1985). Infanticide in a patrilineal society of red colobus monkeys. 20: 323–339. K. (1985). Guinea. T. M.

by Campbell. C. Infanticide by male primates: the sexual selection hypothesis revisited. L. Nature. T. pp. (2000). Trends in Cognitive Sciences. C.R. (1988). The Behavioural Ecology of Humans and Other Mammals (Ed. C.P. Mating conflict in primates: infanticide. Oxford: Blackwell Scientific Publications. M. & Foley.P. (1979).. M. (in press). ).A). 40: 227-233. pp. van Schaik C. B. (1997).B. Tutin. 7(4): 153-156. P. In: Infanticide by males and its implications (Ed.A. van Schaik & Janson. UK.P. Scramble and contest in feeding competition among female long-tailed macaques (Macaca fascicularis).. L. 360: 456. Origins of Human Communication. Tuomi. Production of sperm reduces nematode life-span. C. C. R. 139-179. J. pp 131-150. W. & Mappes. C. Chicago: Adine. by Standen. (1977). K.. & van Schaik.). by C. J.C. & Hare. Call. 50:125–127 145 ..E. In: Sexual selection in primates: new and comparative perspectives (Ed. Journal of Reproductive Fertility. 361. P. Pradhan. (1972). & Nunn.References and Appendix Tomasello. Van Voorhies. (1992).. van Noordwijk M. Parental investment and sexual selection. (2000). In: Infanticide by Males and its Implications (Ed. (2004). (1996). The ecology of social relationships amongst female primates.D Thesis. Behaviour. Tomasello. J. R. G. pp 27–60 Cambridge University Press. The structure and function of the long-distance calls given by male chimpanzees in Gombe National Park. Unpublished Ph. Behavioural Ecology and Sociobiology. J. In: Comparative Socioecology:. (1989).A. Cambridge University Press. C.H.). sexual harassment and female sexuality. Uhlenbroek. van Schaik. 195-218. & van Noordwijk. Behavioural Ecology and Sociobiology. C. pp. Viljoen S. 105: 77-98. 6: 29-38.A.G. M.. Cambridge. van Schaik. Chimpanzees understand psychological states the question is which ones and to what extent. Factors affecting breeding synchronization in an African bush squirrel (Paraxerus cepapi cepapi). Bristol. Mating patterns and Reproductive strategies in a community of wild chimpanzees (Pan troglodytes schweinfurthii). P. Paternity confusion and the ovarian cycles of female primates. by van Schaik. Cambridge. V.M.. P. MIT Press Trivers. Cambridge: Cambridge University Press. C. Agrell. van Schaik C.P& Janson. In: Sexual selection and the decent of man (Ed.H). On the evolutionary stability of female infanticide. by Kappeler. van Schaik.P. (2003). Hodges.

Why do male chimpanzees defend a group range? Animal Behaviour. (2003).G.. Whiten.. Y. D. V. J. 7(7): 237-241. Coalitionary mate guarding by male chimpanzees at Ngogo. Behaviour.K. PhD thesis. Wasser. J. How do apes ape? Learning and Behaviour. 138: 1481-1516. (2002). (2001). Trends Ecol Evol. & Rosenqvist. S. 44:43-55 Watts. Wittig.. Food competition and linear dominance hierarchy among female (Pan troglodytes verus) of the Taï National Park. Williams. J. M. Kibale National Park. W. A. A. Intergroup relations in chimpanzees. Annual Review of Anthropology.. V. and Boesch.are they all sex role reversed. Nishida. Does Participation in Intergroup Conflict Depend on Numerical Assessment. Horner. B. (1998). C.. Nature..J. A. (2002). 56: 513-538. 32: 363-392. Female strategies and the reasons for territoriality in chimpanzees: lessons from three decades of research at Gombe. Trends in Ecology and Evolution. M. D. (2004).A. Behavioural Ecology and Sociobiology. M. and Barash..W. International Journal of Primatology. Wilson. W. L. A. (2005). & Marshall-Pescini. A. Cultures in chimpanzees.. Whiten. S. (1999). A.C and Sherrow. N. Carlis. J. Litchfield. M. C. Primates. & de Waal. Gage. Pipefishes and seahorses . McGrew. Nishida.M.. male prudence and sperm-limited females. and Parker G. 32: 36-52. M. L. & Reynolds.P. Whiten. 17:313–320. G. A. Weddell.. F. Minneapolis: University of Minnesota. V. Wilson. Charting cultural variation in chimpanzees.. (2003). Ahnesjoi. Sugiyama. J.. or Rank for Wild Chimpanzees? Animal Behaviour. Whiten. W. Watts... (2001). Kibale National Park. 61(6): 1203-1216. McGrew. 399: 682.D and Wrangham. T. et al.. E. Mitani. & Pusey. Sperm competition. Conformity to cultural norms of tool use in chimpanzees. Horner. C..References and Appendix Vincent. Reproductive suppression among female mammals: Implications for biomedicine and sexual selection theory. R. Uganda. Oehlert. 437: 737-740. W. T. R. H. Reynolds. 24(4): 847-867 146 ... Goodall. Hauser. Range Location. (1999).. New cases of inter-community infanticide chimpanzees at Ngogo. Goodall. (1983). (1992). V. B. D. Uganda. Nature. J. 68: 523-532. & Wrangham. G. A. (2004). 43: 263-270.. R. Quarterly Review of Biology..P. C. V...... Williams.

Antifeedants. 35(3): 175-183. 111(4): 445-449. de Waal).References and Appendix Wittig. Crockford. Wrangham. W. by B. Behavioural Ecology and Sociobiology.L.W. Wrangham. and Smuts. R.B. K. Wrangham. 75: 262-299. R. Wrangham. (2007).W. W.L. (1997). E.. W. H.M. Nishida. McGrew. Cheney. Science.. G. Demonic males. D. Clark. Smuts. W. Marler.W. K. R. Wrangham. Schwibbe. M. International Journal of Primatology. Chicago: Univ. 77: 774-775. Dietary response of chimpanzees and cercopithecines to seasonal variation in fruit abundance: I. D. B. R. Wrangham. Sex differences in the behavioural ecology of chimpanzees in Gombe National park. Female social relationships and social organization of Kibale Forest chimpanzees. 18(3): 335-368.. 19: 949-970. P. Wrangham. (1987). by T. Seyfarth. & Kaumanns. Tanzania. R.. and Hunt..D (1998). Proceedings of the Royal Society of London B. B. of Chicago Press. (1980). (1987).. (1980).L. R. pp. & Isabirye-Basuta. B. 282. & Peterson. & Cheney.(1979). L. Subtle. R. (1992).M. (2000). M. Cycle Synchrony and Probability of Conception in Female Hamadryas Baboons (Papio-Hamadryas). W. Wrangham & T. R. R. R. An ecological model of female-bonded primate groups. Conklin-Brittain. A. M. C. R. W. R. Wrangham. Zinner. C. D. N.. Evolution of social structure.. Tokyo: University of Tokyo Press. R. New York: Houghton Mifflin. W. Zoo Biology.W. Wikberg. Wrangham. secret female chimpanzees. (1996). American Journal of Physical Anthropology. D. 28:13-31. pp. 8198. Seyfarth. Kinmediated reconciliation substitutes for direct reconciliation in female baboons. Journal of Reproductive Fertility. Zuberbuhler. 6:219-223. 147 . Social Sciences Information. A view on the science: Physical anthropology at the millennium. Behaviour.R190 Zucker. In: Topics in Primatology (Ed. (1994). P. M. Pickford & F. Audience effects. On the evolution of ape social systems. E. In Primate Societies (Ed. Control of intragroup aggression by a captive male orangutan. Struhsaker). W. 274: 1109-1115. 18(5): R189 . T. (2007). Current Biology.

illustrating how the four temporal and spectral parameters were measured from each scream. Appendix Figure 1: A tantrum scream produced by infant Karo of the Sonso community. Owren calculates the average frequency over this formant (Hz).37 and scripts written by M. Here I will briefly outline the acoustic analysis protocol undertaken and the results obtained (see appendix figure 1 and appendix table 1). 3) Mean frequency of the 1st formant: the average frequency of the screams 1st formant band (Hz). B: the peak frequency of the scream (Hz). D: the blue line 148 . Owren (see appendix figure 1): 1) Duration: the length of the individual screams constituting a scream bout (s). A script written by M. mild and tantrum scream acoustic analysis In chapter four I was able to able to replicate the pattern of acoustic differences between mild and severe screams reported by Slocombe and Zuberbuhler (2007). C: the first red dotted line indicates the tracking of the 1st formant. Uganda. Methods I measured victim and tantrum screams along four independent parameters using PRAAT version 4. Furthermore I showed that severe screams and tantrum screams are matched in their acoustic structure. 4) Peak frequency in the call: highest frequency in the fundamental (Hz). 2) Mean pitch of the scream: the scream’s average pitch (indicated by blue line) (Hz).3.References and Appendix Appendix I Severe. Budongo Forest. A: the duration of the scream (s).

T-values revealed a significant difference (indicated by *) only in duration. a script written by M.4 201 1276 1. From this.54 0.25 1064 2.References and Appendix indicates the tracking of the pitch of the scream. Results Peak frequency of fundamental SD 1435. N refers to the number of scream bouts (used as stimuli) that contributed to the mean values reported in each catergory.619* 58.6 215 Appendix table 1: The mean values and SD of four acoustic measures for mild and severe scream conditions.3 0. 149 . or “pitch” of the fundamental (Hz).17 0.149 Mean pitch 62 189.8 2.31 0.15 Duration SD 1404. Mean values of parameters matched the desriptive statistics reported in Slocombe and Zuberbuhler (2007). An identical method is used to calculate pitch in copulation calls (chapter five).45 Mild (N=7) t value 1260.933 1st formant SD Trial type Severe (N=5) 1435.54 38.543 SD 0. though this is highly likely due to the small sample sizes used (N=7 Vs N=5). Owren calculates the mean frequency.

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.